Upload folder using huggingface_hub
Browse files- chat_template.jinja +1 -0
- config.json +13 -0
- model.safetensors +3 -0
- optimizer.pt +3 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- special_tokens_map.json +5 -0
- tokenizer_config.json +38 -0
- trainer_state.json +3034 -0
- training_args.bin +3 -0
chat_template.jinja
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{# -------------------------------Optional: Define available tools------------------------------- #}{%- if tools %}{%- for tool in tools %}{{- "\x11" -}} {# Start Of Tool Definition #}{{- "\x0E" -}} {# Start Of Attention Block #}{{- tool['function']['name'] + '\n' -}}{# Loop through all argument names and their descriptions #}{%- for argument in tool['function']['parameters']['properties'] %}{{- argument + ': ' + tool['function']['parameters']['properties'][argument]['description'] + '\n' -}}{%- endfor %}{{- '\x0F' -}} {# End Of Attention Block #}{%- endfor %}{%- endif %}{# -------------------------------Main conversation message loop------------------------------- #}{% for message in messages %}{{- "\x01" -}} {# Start Of Text Block #}{# Print the role tag, e.g. user or assistant #}{{- message.role + "\n" -}}{# If not the "assistant", we wrap with an attention block #}{% if message.role != "assistant" %}{{- "\x0E" -}}{% endif %}{# If the message contains normal content, print it #}{% if message.content %}{{- message.content -}}{% endif %}{# If the assistant called any tools, print those tool calls #}{% if message.tool_calls %}{% for call in message.tool_calls %}{{- "\x1A" -}} {# Start Of Tool Call #}{{- call.type + " " + call.function.name+ "(" + call.function.arguments | tojson + ")" -}}{{- "\x1B" -}} {# End Of Tool Call #}{% endfor %}{% endif %}{# If not the "assistant", we wrap with an attention block #}{% if message.role != "assistant" %}{{- "\x0F" -}}{% endif %}{{- "\x17" -}} {# End Of Text Block #}{{- "\n" -}} {# Newline after each message #}{% endfor %}{# -------------------------------Add a final assistant prompt marker to continue generation------------------------------- #}{%- if add_generation_prompt %}{{- "\x01assistant\n" -}}{%- endif %}
|
config.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"CharacterCausalLMWrapper"
|
| 4 |
+
],
|
| 5 |
+
"base_model_name_or_path": "sign/utf8-lm-tiny",
|
| 6 |
+
"bos_token_id": 2,
|
| 7 |
+
"dtype": "float32",
|
| 8 |
+
"eos_token_id": 3,
|
| 9 |
+
"model_type": "character_causal_lm",
|
| 10 |
+
"num_bytes": 2,
|
| 11 |
+
"pad_token_id": 0,
|
| 12 |
+
"transformers_version": "4.57.3"
|
| 13 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:58440bfed945c700604bc2bd146247b87700d59e81fe3d443a095662bb6c81f4
|
| 3 |
+
size 13133616
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6c9418bea7e96f5109e2a38ca0fca089ac2a73d5f346f8fb663ea47f1dbf233f
|
| 3 |
+
size 25228427
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:61c19bab1174704a4a4441475683bf1270277af15d2e2c95e964789128e482c4
|
| 3 |
+
size 14645
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d048e19eb3a548b350d1f6f2b3bdac8b5eebdfd4fc56882e1e6d68412775b9cd
|
| 3 |
+
size 1465
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,5 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": "\u0002",
|
| 3 |
+
"eos_token": "\u0003",
|
| 4 |
+
"pad_token": "\u0000"
|
| 5 |
+
}
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,38 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"added_tokens_decoder": {
|
| 3 |
+
"0": {
|
| 4 |
+
"content": "\u0000",
|
| 5 |
+
"lstrip": false,
|
| 6 |
+
"normalized": false,
|
| 7 |
+
"rstrip": false,
|
| 8 |
+
"single_word": false,
|
| 9 |
+
"special": true
|
| 10 |
+
},
|
| 11 |
+
"2": {
|
| 12 |
+
"content": "\u0002",
|
| 13 |
+
"lstrip": false,
|
| 14 |
+
"normalized": false,
|
| 15 |
+
"rstrip": false,
|
| 16 |
+
"single_word": false,
|
| 17 |
+
"special": true
|
| 18 |
+
},
|
| 19 |
+
"3": {
|
| 20 |
+
"content": "\u0003",
|
| 21 |
+
"lstrip": false,
|
| 22 |
+
"normalized": false,
|
| 23 |
+
"rstrip": false,
|
| 24 |
+
"single_word": false,
|
| 25 |
+
"special": true
|
| 26 |
+
}
|
| 27 |
+
},
|
| 28 |
+
"bos_token": "\u0002",
|
| 29 |
+
"bos_token_id": 2,
|
| 30 |
+
"clean_up_tokenization_spaces": false,
|
| 31 |
+
"eos_token": "\u0003",
|
| 32 |
+
"eos_token_id": 3,
|
| 33 |
+
"extra_special_tokens": {},
|
| 34 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 35 |
+
"pad_token": "\u0000",
|
| 36 |
+
"pad_token_id": 0,
|
| 37 |
+
"tokenizer_class": "UTF16Tokenizer"
|
| 38 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,3034 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 0.3,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 30000,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.001,
|
| 14 |
+
"grad_norm": 20.30853843688965,
|
| 15 |
+
"learning_rate": 2.97e-05,
|
| 16 |
+
"loss": 6.3768,
|
| 17 |
+
"num_input_tokens_seen": 6553600,
|
| 18 |
+
"step": 100,
|
| 19 |
+
"train_runtime": 67.7249,
|
| 20 |
+
"train_tokens_per_second": 96767.992
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.002,
|
| 24 |
+
"grad_norm": 12.960590362548828,
|
| 25 |
+
"learning_rate": 5.97e-05,
|
| 26 |
+
"loss": 1.5751,
|
| 27 |
+
"num_input_tokens_seen": 13107200,
|
| 28 |
+
"step": 200,
|
| 29 |
+
"train_runtime": 121.8001,
|
| 30 |
+
"train_tokens_per_second": 107612.389
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.003,
|
| 34 |
+
"grad_norm": 4.234757900238037,
|
| 35 |
+
"learning_rate": 8.969999999999998e-05,
|
| 36 |
+
"loss": 1.3419,
|
| 37 |
+
"num_input_tokens_seen": 19660800,
|
| 38 |
+
"step": 300,
|
| 39 |
+
"train_runtime": 175.7236,
|
| 40 |
+
"train_tokens_per_second": 111884.791
|
| 41 |
+
},
|
| 42 |
+
{
|
| 43 |
+
"epoch": 0.004,
|
| 44 |
+
"grad_norm": 2.8189597129821777,
|
| 45 |
+
"learning_rate": 0.0001197,
|
| 46 |
+
"loss": 1.1467,
|
| 47 |
+
"num_input_tokens_seen": 26214400,
|
| 48 |
+
"step": 400,
|
| 49 |
+
"train_runtime": 229.814,
|
| 50 |
+
"train_tokens_per_second": 114067.895
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.005,
|
| 54 |
+
"grad_norm": 4.593051910400391,
|
| 55 |
+
"learning_rate": 0.00014969999999999998,
|
| 56 |
+
"loss": 1.02,
|
| 57 |
+
"num_input_tokens_seen": 32768000,
|
| 58 |
+
"step": 500,
|
| 59 |
+
"train_runtime": 288.5437,
|
| 60 |
+
"train_tokens_per_second": 113563.389
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 0.006,
|
| 64 |
+
"grad_norm": 1.745302677154541,
|
| 65 |
+
"learning_rate": 0.00017969999999999998,
|
| 66 |
+
"loss": 0.9381,
|
| 67 |
+
"num_input_tokens_seen": 39321600,
|
| 68 |
+
"step": 600,
|
| 69 |
+
"train_runtime": 343.6994,
|
| 70 |
+
"train_tokens_per_second": 114406.961
|
| 71 |
+
},
|
| 72 |
+
{
|
| 73 |
+
"epoch": 0.007,
|
| 74 |
+
"grad_norm": 2.180215358734131,
|
| 75 |
+
"learning_rate": 0.00020969999999999997,
|
| 76 |
+
"loss": 0.8912,
|
| 77 |
+
"num_input_tokens_seen": 45875200,
|
| 78 |
+
"step": 700,
|
| 79 |
+
"train_runtime": 398.5625,
|
| 80 |
+
"train_tokens_per_second": 115101.641
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.008,
|
| 84 |
+
"grad_norm": 1.5782365798950195,
|
| 85 |
+
"learning_rate": 0.0002397,
|
| 86 |
+
"loss": 0.8525,
|
| 87 |
+
"num_input_tokens_seen": 52428800,
|
| 88 |
+
"step": 800,
|
| 89 |
+
"train_runtime": 453.9393,
|
| 90 |
+
"train_tokens_per_second": 115497.369
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 0.009,
|
| 94 |
+
"grad_norm": 1.306846022605896,
|
| 95 |
+
"learning_rate": 0.0002697,
|
| 96 |
+
"loss": 0.8411,
|
| 97 |
+
"num_input_tokens_seen": 58982400,
|
| 98 |
+
"step": 900,
|
| 99 |
+
"train_runtime": 508.8336,
|
| 100 |
+
"train_tokens_per_second": 115916.867
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.01,
|
| 104 |
+
"grad_norm": 0.9485424160957336,
|
| 105 |
+
"learning_rate": 0.00029969999999999997,
|
| 106 |
+
"loss": 0.8207,
|
| 107 |
+
"num_input_tokens_seen": 65536000,
|
| 108 |
+
"step": 1000,
|
| 109 |
+
"train_runtime": 569.5888,
|
| 110 |
+
"train_tokens_per_second": 115058.451
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 0.011,
|
| 114 |
+
"grad_norm": 0.8611180186271667,
|
| 115 |
+
"learning_rate": 0.00029999925978027874,
|
| 116 |
+
"loss": 0.8158,
|
| 117 |
+
"num_input_tokens_seen": 72089600,
|
| 118 |
+
"step": 1100,
|
| 119 |
+
"train_runtime": 623.5912,
|
| 120 |
+
"train_tokens_per_second": 115603.946
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 0.012,
|
| 124 |
+
"grad_norm": 0.8997637033462524,
|
| 125 |
+
"learning_rate": 0.0002999970091452017,
|
| 126 |
+
"loss": 0.7948,
|
| 127 |
+
"num_input_tokens_seen": 78643200,
|
| 128 |
+
"step": 1200,
|
| 129 |
+
"train_runtime": 679.5203,
|
| 130 |
+
"train_tokens_per_second": 115733.415
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 0.013,
|
| 134 |
+
"grad_norm": 0.6279132962226868,
|
| 135 |
+
"learning_rate": 0.00029999324804190795,
|
| 136 |
+
"loss": 0.7852,
|
| 137 |
+
"num_input_tokens_seen": 85196800,
|
| 138 |
+
"step": 1300,
|
| 139 |
+
"train_runtime": 739.7161,
|
| 140 |
+
"train_tokens_per_second": 115174.993
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 0.014,
|
| 144 |
+
"grad_norm": 0.5959413051605225,
|
| 145 |
+
"learning_rate": 0.0002999879765082716,
|
| 146 |
+
"loss": 0.7736,
|
| 147 |
+
"num_input_tokens_seen": 91750400,
|
| 148 |
+
"step": 1400,
|
| 149 |
+
"train_runtime": 794.2065,
|
| 150 |
+
"train_tokens_per_second": 115524.612
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.015,
|
| 154 |
+
"grad_norm": 0.6557334065437317,
|
| 155 |
+
"learning_rate": 0.000299981194597377,
|
| 156 |
+
"loss": 0.7631,
|
| 157 |
+
"num_input_tokens_seen": 98304000,
|
| 158 |
+
"step": 1500,
|
| 159 |
+
"train_runtime": 847.9465,
|
| 160 |
+
"train_tokens_per_second": 115931.844
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"epoch": 0.016,
|
| 164 |
+
"grad_norm": 0.7466850876808167,
|
| 165 |
+
"learning_rate": 0.0002999729023775179,
|
| 166 |
+
"loss": 0.7564,
|
| 167 |
+
"num_input_tokens_seen": 104857600,
|
| 168 |
+
"step": 1600,
|
| 169 |
+
"train_runtime": 907.419,
|
| 170 |
+
"train_tokens_per_second": 115555.874
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.017,
|
| 174 |
+
"grad_norm": 0.6613496541976929,
|
| 175 |
+
"learning_rate": 0.0002999630999321969,
|
| 176 |
+
"loss": 0.7664,
|
| 177 |
+
"num_input_tokens_seen": 111411200,
|
| 178 |
+
"step": 1700,
|
| 179 |
+
"train_runtime": 957.6918,
|
| 180 |
+
"train_tokens_per_second": 116333.047
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"epoch": 0.018,
|
| 184 |
+
"grad_norm": 0.49886658787727356,
|
| 185 |
+
"learning_rate": 0.00029995178736012443,
|
| 186 |
+
"loss": 0.7577,
|
| 187 |
+
"num_input_tokens_seen": 117964800,
|
| 188 |
+
"step": 1800,
|
| 189 |
+
"train_runtime": 1017.5123,
|
| 190 |
+
"train_tokens_per_second": 115934.516
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 0.019,
|
| 194 |
+
"grad_norm": 0.5621941089630127,
|
| 195 |
+
"learning_rate": 0.0002999389647752181,
|
| 196 |
+
"loss": 0.743,
|
| 197 |
+
"num_input_tokens_seen": 124518400,
|
| 198 |
+
"step": 1900,
|
| 199 |
+
"train_runtime": 1070.9552,
|
| 200 |
+
"train_tokens_per_second": 116268.538
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 0.02,
|
| 204 |
+
"grad_norm": 0.46181556582450867,
|
| 205 |
+
"learning_rate": 0.00029992463230660104,
|
| 206 |
+
"loss": 0.7429,
|
| 207 |
+
"num_input_tokens_seen": 131072000,
|
| 208 |
+
"step": 2000,
|
| 209 |
+
"train_runtime": 1130.6338,
|
| 210 |
+
"train_tokens_per_second": 115927.894
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"epoch": 0.021,
|
| 214 |
+
"grad_norm": 0.4571639895439148,
|
| 215 |
+
"learning_rate": 0.00029990879009860117,
|
| 216 |
+
"loss": 0.7349,
|
| 217 |
+
"num_input_tokens_seen": 137625600,
|
| 218 |
+
"step": 2100,
|
| 219 |
+
"train_runtime": 1185.0559,
|
| 220 |
+
"train_tokens_per_second": 116134.269
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.022,
|
| 224 |
+
"grad_norm": 0.5650346875190735,
|
| 225 |
+
"learning_rate": 0.0002998914383107493,
|
| 226 |
+
"loss": 0.7373,
|
| 227 |
+
"num_input_tokens_seen": 144179200,
|
| 228 |
+
"step": 2200,
|
| 229 |
+
"train_runtime": 1239.4558,
|
| 230 |
+
"train_tokens_per_second": 116324.603
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 0.023,
|
| 234 |
+
"grad_norm": 0.5640336871147156,
|
| 235 |
+
"learning_rate": 0.0002998725771177778,
|
| 236 |
+
"loss": 0.7383,
|
| 237 |
+
"num_input_tokens_seen": 150732800,
|
| 238 |
+
"step": 2300,
|
| 239 |
+
"train_runtime": 1294.8284,
|
| 240 |
+
"train_tokens_per_second": 116411.407
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.024,
|
| 244 |
+
"grad_norm": 0.4288141429424286,
|
| 245 |
+
"learning_rate": 0.00029985220670961847,
|
| 246 |
+
"loss": 0.7256,
|
| 247 |
+
"num_input_tokens_seen": 157286400,
|
| 248 |
+
"step": 2400,
|
| 249 |
+
"train_runtime": 1353.9338,
|
| 250 |
+
"train_tokens_per_second": 116169.935
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 0.025,
|
| 254 |
+
"grad_norm": 0.7532113194465637,
|
| 255 |
+
"learning_rate": 0.0002998303272914014,
|
| 256 |
+
"loss": 0.7301,
|
| 257 |
+
"num_input_tokens_seen": 163840000,
|
| 258 |
+
"step": 2500,
|
| 259 |
+
"train_runtime": 1408.3918,
|
| 260 |
+
"train_tokens_per_second": 116331.266
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 0.026,
|
| 264 |
+
"grad_norm": 0.42602330446243286,
|
| 265 |
+
"learning_rate": 0.00029980693908345185,
|
| 266 |
+
"loss": 0.7238,
|
| 267 |
+
"num_input_tokens_seen": 170393600,
|
| 268 |
+
"step": 2600,
|
| 269 |
+
"train_runtime": 1461.9698,
|
| 270 |
+
"train_tokens_per_second": 116550.699
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 0.027,
|
| 274 |
+
"grad_norm": 0.655644953250885,
|
| 275 |
+
"learning_rate": 0.00029978204232128895,
|
| 276 |
+
"loss": 0.7213,
|
| 277 |
+
"num_input_tokens_seen": 176947200,
|
| 278 |
+
"step": 2700,
|
| 279 |
+
"train_runtime": 1523.0525,
|
| 280 |
+
"train_tokens_per_second": 116179.321
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"epoch": 0.028,
|
| 284 |
+
"grad_norm": 0.3962925970554352,
|
| 285 |
+
"learning_rate": 0.0002997556372556227,
|
| 286 |
+
"loss": 0.7238,
|
| 287 |
+
"num_input_tokens_seen": 183500800,
|
| 288 |
+
"step": 2800,
|
| 289 |
+
"train_runtime": 1577.6914,
|
| 290 |
+
"train_tokens_per_second": 116309.692
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.029,
|
| 294 |
+
"grad_norm": 0.44488221406936646,
|
| 295 |
+
"learning_rate": 0.0002997277241523519,
|
| 296 |
+
"loss": 0.7218,
|
| 297 |
+
"num_input_tokens_seen": 190054400,
|
| 298 |
+
"step": 2900,
|
| 299 |
+
"train_runtime": 1631.4556,
|
| 300 |
+
"train_tokens_per_second": 116493.763
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"epoch": 0.03,
|
| 304 |
+
"grad_norm": 0.3886267840862274,
|
| 305 |
+
"learning_rate": 0.00029969830329256125,
|
| 306 |
+
"loss": 0.7091,
|
| 307 |
+
"num_input_tokens_seen": 196608000,
|
| 308 |
+
"step": 3000,
|
| 309 |
+
"train_runtime": 1686.3399,
|
| 310 |
+
"train_tokens_per_second": 116588.595
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.031,
|
| 314 |
+
"grad_norm": 0.46353062987327576,
|
| 315 |
+
"learning_rate": 0.00029966737497251836,
|
| 316 |
+
"loss": 0.7132,
|
| 317 |
+
"num_input_tokens_seen": 203161600,
|
| 318 |
+
"step": 3100,
|
| 319 |
+
"train_runtime": 1745.3481,
|
| 320 |
+
"train_tokens_per_second": 116401.765
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 0.032,
|
| 324 |
+
"grad_norm": 0.45476260781288147,
|
| 325 |
+
"learning_rate": 0.0002996349395036711,
|
| 326 |
+
"loss": 0.7099,
|
| 327 |
+
"num_input_tokens_seen": 209715200,
|
| 328 |
+
"step": 3200,
|
| 329 |
+
"train_runtime": 1799.8354,
|
| 330 |
+
"train_tokens_per_second": 116519.096
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 0.033,
|
| 334 |
+
"grad_norm": 0.5311458706855774,
|
| 335 |
+
"learning_rate": 0.00029960099721264435,
|
| 336 |
+
"loss": 0.7162,
|
| 337 |
+
"num_input_tokens_seen": 216268800,
|
| 338 |
+
"step": 3300,
|
| 339 |
+
"train_runtime": 1854.9316,
|
| 340 |
+
"train_tokens_per_second": 116591.252
|
| 341 |
+
},
|
| 342 |
+
{
|
| 343 |
+
"epoch": 0.034,
|
| 344 |
+
"grad_norm": 0.4546497166156769,
|
| 345 |
+
"learning_rate": 0.0002995655484412365,
|
| 346 |
+
"loss": 0.7046,
|
| 347 |
+
"num_input_tokens_seen": 222822400,
|
| 348 |
+
"step": 3400,
|
| 349 |
+
"train_runtime": 1909.1868,
|
| 350 |
+
"train_tokens_per_second": 116710.63
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 0.035,
|
| 354 |
+
"grad_norm": 0.6349391937255859,
|
| 355 |
+
"learning_rate": 0.00029952859354641636,
|
| 356 |
+
"loss": 0.7073,
|
| 357 |
+
"num_input_tokens_seen": 229376000,
|
| 358 |
+
"step": 3500,
|
| 359 |
+
"train_runtime": 1969.0427,
|
| 360 |
+
"train_tokens_per_second": 116491.125
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.036,
|
| 364 |
+
"grad_norm": 0.493557870388031,
|
| 365 |
+
"learning_rate": 0.00029949013290031924,
|
| 366 |
+
"loss": 0.7021,
|
| 367 |
+
"num_input_tokens_seen": 235929600,
|
| 368 |
+
"step": 3600,
|
| 369 |
+
"train_runtime": 2023.4924,
|
| 370 |
+
"train_tokens_per_second": 116595.251
|
| 371 |
+
},
|
| 372 |
+
{
|
| 373 |
+
"epoch": 0.037,
|
| 374 |
+
"grad_norm": 0.3123897314071655,
|
| 375 |
+
"learning_rate": 0.00029945016689024353,
|
| 376 |
+
"loss": 0.696,
|
| 377 |
+
"num_input_tokens_seen": 242483200,
|
| 378 |
+
"step": 3700,
|
| 379 |
+
"train_runtime": 2077.3277,
|
| 380 |
+
"train_tokens_per_second": 116728.432
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.038,
|
| 384 |
+
"grad_norm": 0.43099701404571533,
|
| 385 |
+
"learning_rate": 0.0002994086959186464,
|
| 386 |
+
"loss": 0.6997,
|
| 387 |
+
"num_input_tokens_seen": 249036800,
|
| 388 |
+
"step": 3800,
|
| 389 |
+
"train_runtime": 2132.1624,
|
| 390 |
+
"train_tokens_per_second": 116800.109
|
| 391 |
+
},
|
| 392 |
+
{
|
| 393 |
+
"epoch": 0.039,
|
| 394 |
+
"grad_norm": 0.3274790346622467,
|
| 395 |
+
"learning_rate": 0.00029936572040314014,
|
| 396 |
+
"loss": 0.7034,
|
| 397 |
+
"num_input_tokens_seen": 255590400,
|
| 398 |
+
"step": 3900,
|
| 399 |
+
"train_runtime": 2193.0795,
|
| 400 |
+
"train_tokens_per_second": 116544.063
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 0.04,
|
| 404 |
+
"grad_norm": 0.33723658323287964,
|
| 405 |
+
"learning_rate": 0.0002993212407764877,
|
| 406 |
+
"loss": 0.6977,
|
| 407 |
+
"num_input_tokens_seen": 262144000,
|
| 408 |
+
"step": 4000,
|
| 409 |
+
"train_runtime": 2246.9274,
|
| 410 |
+
"train_tokens_per_second": 116667.767
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 0.041,
|
| 414 |
+
"grad_norm": 0.42106112837791443,
|
| 415 |
+
"learning_rate": 0.00029927525748659834,
|
| 416 |
+
"loss": 0.7078,
|
| 417 |
+
"num_input_tokens_seen": 268697600,
|
| 418 |
+
"step": 4100,
|
| 419 |
+
"train_runtime": 2302.3386,
|
| 420 |
+
"train_tokens_per_second": 116706.38
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"epoch": 0.042,
|
| 424 |
+
"grad_norm": 0.3567107915878296,
|
| 425 |
+
"learning_rate": 0.0002992277709965234,
|
| 426 |
+
"loss": 0.6967,
|
| 427 |
+
"num_input_tokens_seen": 275251200,
|
| 428 |
+
"step": 4200,
|
| 429 |
+
"train_runtime": 2356.6313,
|
| 430 |
+
"train_tokens_per_second": 116798.585
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.043,
|
| 434 |
+
"grad_norm": 0.4269777834415436,
|
| 435 |
+
"learning_rate": 0.0002991787817844513,
|
| 436 |
+
"loss": 0.6976,
|
| 437 |
+
"num_input_tokens_seen": 281804800,
|
| 438 |
+
"step": 4300,
|
| 439 |
+
"train_runtime": 2417.993,
|
| 440 |
+
"train_tokens_per_second": 116544.92
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 0.044,
|
| 444 |
+
"grad_norm": 0.4035900831222534,
|
| 445 |
+
"learning_rate": 0.0002991282903437028,
|
| 446 |
+
"loss": 0.6915,
|
| 447 |
+
"num_input_tokens_seen": 288358400,
|
| 448 |
+
"step": 4400,
|
| 449 |
+
"train_runtime": 2471.1514,
|
| 450 |
+
"train_tokens_per_second": 116689.896
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.045,
|
| 454 |
+
"grad_norm": 0.3197948634624481,
|
| 455 |
+
"learning_rate": 0.0002990762971827262,
|
| 456 |
+
"loss": 0.6905,
|
| 457 |
+
"num_input_tokens_seen": 294912000,
|
| 458 |
+
"step": 4500,
|
| 459 |
+
"train_runtime": 2525.6119,
|
| 460 |
+
"train_tokens_per_second": 116768.533
|
| 461 |
+
},
|
| 462 |
+
{
|
| 463 |
+
"epoch": 0.046,
|
| 464 |
+
"grad_norm": 0.3118247389793396,
|
| 465 |
+
"learning_rate": 0.00029902280282509197,
|
| 466 |
+
"loss": 0.6959,
|
| 467 |
+
"num_input_tokens_seen": 301465600,
|
| 468 |
+
"step": 4600,
|
| 469 |
+
"train_runtime": 2587.1477,
|
| 470 |
+
"train_tokens_per_second": 116524.311
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 0.047,
|
| 474 |
+
"grad_norm": 0.38194310665130615,
|
| 475 |
+
"learning_rate": 0.0002989678078094878,
|
| 476 |
+
"loss": 0.6812,
|
| 477 |
+
"num_input_tokens_seen": 308019200,
|
| 478 |
+
"step": 4700,
|
| 479 |
+
"train_runtime": 2645.2255,
|
| 480 |
+
"train_tokens_per_second": 116443.456
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 0.048,
|
| 484 |
+
"grad_norm": 0.5399206876754761,
|
| 485 |
+
"learning_rate": 0.00029891131268971284,
|
| 486 |
+
"loss": 0.6843,
|
| 487 |
+
"num_input_tokens_seen": 314572800,
|
| 488 |
+
"step": 4800,
|
| 489 |
+
"train_runtime": 2702.2839,
|
| 490 |
+
"train_tokens_per_second": 116409.974
|
| 491 |
+
},
|
| 492 |
+
{
|
| 493 |
+
"epoch": 0.049,
|
| 494 |
+
"grad_norm": 0.3325801491737366,
|
| 495 |
+
"learning_rate": 0.0002988533180346723,
|
| 496 |
+
"loss": 0.6803,
|
| 497 |
+
"num_input_tokens_seen": 321126400,
|
| 498 |
+
"step": 4900,
|
| 499 |
+
"train_runtime": 2761.723,
|
| 500 |
+
"train_tokens_per_second": 116277.558
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.05,
|
| 504 |
+
"grad_norm": 0.41960838437080383,
|
| 505 |
+
"learning_rate": 0.0002987938244283717,
|
| 506 |
+
"loss": 0.6772,
|
| 507 |
+
"num_input_tokens_seen": 327680000,
|
| 508 |
+
"step": 5000,
|
| 509 |
+
"train_runtime": 2818.6212,
|
| 510 |
+
"train_tokens_per_second": 116255.424
|
| 511 |
+
},
|
| 512 |
+
{
|
| 513 |
+
"epoch": 0.051,
|
| 514 |
+
"grad_norm": 0.40982258319854736,
|
| 515 |
+
"learning_rate": 0.00029873283246991105,
|
| 516 |
+
"loss": 0.6861,
|
| 517 |
+
"num_input_tokens_seen": 334233600,
|
| 518 |
+
"step": 5100,
|
| 519 |
+
"train_runtime": 2873.5528,
|
| 520 |
+
"train_tokens_per_second": 116313.715
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.052,
|
| 524 |
+
"grad_norm": 0.31062355637550354,
|
| 525 |
+
"learning_rate": 0.0002986703427734787,
|
| 526 |
+
"loss": 0.6852,
|
| 527 |
+
"num_input_tokens_seen": 340787200,
|
| 528 |
+
"step": 5200,
|
| 529 |
+
"train_runtime": 2929.1895,
|
| 530 |
+
"train_tokens_per_second": 116341.805
|
| 531 |
+
},
|
| 532 |
+
{
|
| 533 |
+
"epoch": 0.053,
|
| 534 |
+
"grad_norm": 0.3270035684108734,
|
| 535 |
+
"learning_rate": 0.00029860635596834517,
|
| 536 |
+
"loss": 0.6854,
|
| 537 |
+
"num_input_tokens_seen": 347340800,
|
| 538 |
+
"step": 5300,
|
| 539 |
+
"train_runtime": 2983.4815,
|
| 540 |
+
"train_tokens_per_second": 116421.301
|
| 541 |
+
},
|
| 542 |
+
{
|
| 543 |
+
"epoch": 0.054,
|
| 544 |
+
"grad_norm": 0.3153812289237976,
|
| 545 |
+
"learning_rate": 0.0002985408726988569,
|
| 546 |
+
"loss": 0.6826,
|
| 547 |
+
"num_input_tokens_seen": 353894400,
|
| 548 |
+
"step": 5400,
|
| 549 |
+
"train_runtime": 3042.2104,
|
| 550 |
+
"train_tokens_per_second": 116328.049
|
| 551 |
+
},
|
| 552 |
+
{
|
| 553 |
+
"epoch": 0.055,
|
| 554 |
+
"grad_norm": 0.328072190284729,
|
| 555 |
+
"learning_rate": 0.0002984738936244296,
|
| 556 |
+
"loss": 0.6796,
|
| 557 |
+
"num_input_tokens_seen": 360448000,
|
| 558 |
+
"step": 5500,
|
| 559 |
+
"train_runtime": 3096.1295,
|
| 560 |
+
"train_tokens_per_second": 116418.904
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 0.056,
|
| 564 |
+
"grad_norm": 0.32828816771507263,
|
| 565 |
+
"learning_rate": 0.0002984054194195419,
|
| 566 |
+
"loss": 0.6927,
|
| 567 |
+
"num_input_tokens_seen": 367001600,
|
| 568 |
+
"step": 5600,
|
| 569 |
+
"train_runtime": 3151.7383,
|
| 570 |
+
"train_tokens_per_second": 116444.184
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.057,
|
| 574 |
+
"grad_norm": 0.3481093645095825,
|
| 575 |
+
"learning_rate": 0.0002983354507737283,
|
| 576 |
+
"loss": 0.6876,
|
| 577 |
+
"num_input_tokens_seen": 373555200,
|
| 578 |
+
"step": 5700,
|
| 579 |
+
"train_runtime": 3207.4,
|
| 580 |
+
"train_tokens_per_second": 116466.67
|
| 581 |
+
},
|
| 582 |
+
{
|
| 583 |
+
"epoch": 0.058,
|
| 584 |
+
"grad_norm": 0.4054113030433655,
|
| 585 |
+
"learning_rate": 0.00029826398839157215,
|
| 586 |
+
"loss": 0.6868,
|
| 587 |
+
"num_input_tokens_seen": 380108800,
|
| 588 |
+
"step": 5800,
|
| 589 |
+
"train_runtime": 3267.0971,
|
| 590 |
+
"train_tokens_per_second": 116344.508
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.059,
|
| 594 |
+
"grad_norm": 0.3671651780605316,
|
| 595 |
+
"learning_rate": 0.000298191032992699,
|
| 596 |
+
"loss": 0.6818,
|
| 597 |
+
"num_input_tokens_seen": 386662400,
|
| 598 |
+
"step": 5900,
|
| 599 |
+
"train_runtime": 3320.6147,
|
| 600 |
+
"train_tokens_per_second": 116443.019
|
| 601 |
+
},
|
| 602 |
+
{
|
| 603 |
+
"epoch": 0.06,
|
| 604 |
+
"grad_norm": 0.2815225422382355,
|
| 605 |
+
"learning_rate": 0.0002981165853117688,
|
| 606 |
+
"loss": 0.6735,
|
| 607 |
+
"num_input_tokens_seen": 393216000,
|
| 608 |
+
"step": 6000,
|
| 609 |
+
"train_runtime": 3380.4372,
|
| 610 |
+
"train_tokens_per_second": 116321.049
|
| 611 |
+
},
|
| 612 |
+
{
|
| 613 |
+
"epoch": 0.061,
|
| 614 |
+
"grad_norm": 0.5300637483596802,
|
| 615 |
+
"learning_rate": 0.000298040646098469,
|
| 616 |
+
"loss": 0.6792,
|
| 617 |
+
"num_input_tokens_seen": 399769600,
|
| 618 |
+
"step": 6100,
|
| 619 |
+
"train_runtime": 3434.9665,
|
| 620 |
+
"train_tokens_per_second": 116382.388
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"epoch": 0.062,
|
| 624 |
+
"grad_norm": 0.28088632225990295,
|
| 625 |
+
"learning_rate": 0.0002979632161175064,
|
| 626 |
+
"loss": 0.6763,
|
| 627 |
+
"num_input_tokens_seen": 406323200,
|
| 628 |
+
"step": 6200,
|
| 629 |
+
"train_runtime": 3490.2084,
|
| 630 |
+
"train_tokens_per_second": 116418.034
|
| 631 |
+
},
|
| 632 |
+
{
|
| 633 |
+
"epoch": 0.063,
|
| 634 |
+
"grad_norm": 0.5535532236099243,
|
| 635 |
+
"learning_rate": 0.0002978842961486003,
|
| 636 |
+
"loss": 0.6769,
|
| 637 |
+
"num_input_tokens_seen": 412876800,
|
| 638 |
+
"step": 6300,
|
| 639 |
+
"train_runtime": 3587.6493,
|
| 640 |
+
"train_tokens_per_second": 115082.821
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.064,
|
| 644 |
+
"grad_norm": 0.3341565728187561,
|
| 645 |
+
"learning_rate": 0.0002978038869864738,
|
| 646 |
+
"loss": 0.6734,
|
| 647 |
+
"num_input_tokens_seen": 419430400,
|
| 648 |
+
"step": 6400,
|
| 649 |
+
"train_runtime": 3650.2458,
|
| 650 |
+
"train_tokens_per_second": 114904.702
|
| 651 |
+
},
|
| 652 |
+
{
|
| 653 |
+
"epoch": 0.065,
|
| 654 |
+
"grad_norm": 0.34299543499946594,
|
| 655 |
+
"learning_rate": 0.0002977219894408463,
|
| 656 |
+
"loss": 0.671,
|
| 657 |
+
"num_input_tokens_seen": 425984000,
|
| 658 |
+
"step": 6500,
|
| 659 |
+
"train_runtime": 3704.075,
|
| 660 |
+
"train_tokens_per_second": 115004.151
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.066,
|
| 664 |
+
"grad_norm": 0.3468911349773407,
|
| 665 |
+
"learning_rate": 0.0002976386043364251,
|
| 666 |
+
"loss": 0.6799,
|
| 667 |
+
"num_input_tokens_seen": 432537600,
|
| 668 |
+
"step": 6600,
|
| 669 |
+
"train_runtime": 3758.2332,
|
| 670 |
+
"train_tokens_per_second": 115090.677
|
| 671 |
+
},
|
| 672 |
+
{
|
| 673 |
+
"epoch": 0.067,
|
| 674 |
+
"grad_norm": 0.29494985938072205,
|
| 675 |
+
"learning_rate": 0.00029755373251289733,
|
| 676 |
+
"loss": 0.6836,
|
| 677 |
+
"num_input_tokens_seen": 439091200,
|
| 678 |
+
"step": 6700,
|
| 679 |
+
"train_runtime": 3818.8992,
|
| 680 |
+
"train_tokens_per_second": 114978.475
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"epoch": 0.068,
|
| 684 |
+
"grad_norm": 0.43209826946258545,
|
| 685 |
+
"learning_rate": 0.0002974673748249213,
|
| 686 |
+
"loss": 0.6721,
|
| 687 |
+
"num_input_tokens_seen": 445644800,
|
| 688 |
+
"step": 6800,
|
| 689 |
+
"train_runtime": 3873.8852,
|
| 690 |
+
"train_tokens_per_second": 115038.204
|
| 691 |
+
},
|
| 692 |
+
{
|
| 693 |
+
"epoch": 0.069,
|
| 694 |
+
"grad_norm": 0.36356812715530396,
|
| 695 |
+
"learning_rate": 0.00029737953214211804,
|
| 696 |
+
"loss": 0.6742,
|
| 697 |
+
"num_input_tokens_seen": 452198400,
|
| 698 |
+
"step": 6900,
|
| 699 |
+
"train_runtime": 3928.8659,
|
| 700 |
+
"train_tokens_per_second": 115096.42
|
| 701 |
+
},
|
| 702 |
+
{
|
| 703 |
+
"epoch": 0.07,
|
| 704 |
+
"grad_norm": 0.4425353705883026,
|
| 705 |
+
"learning_rate": 0.0002972902053490623,
|
| 706 |
+
"loss": 0.6691,
|
| 707 |
+
"num_input_tokens_seen": 458752000,
|
| 708 |
+
"step": 7000,
|
| 709 |
+
"train_runtime": 3982.4622,
|
| 710 |
+
"train_tokens_per_second": 115193.058
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 0.071,
|
| 714 |
+
"grad_norm": 0.3378838300704956,
|
| 715 |
+
"learning_rate": 0.00029719939534527393,
|
| 716 |
+
"loss": 0.6817,
|
| 717 |
+
"num_input_tokens_seen": 465305600,
|
| 718 |
+
"step": 7100,
|
| 719 |
+
"train_runtime": 4038.9423,
|
| 720 |
+
"train_tokens_per_second": 115204.814
|
| 721 |
+
},
|
| 722 |
+
{
|
| 723 |
+
"epoch": 0.072,
|
| 724 |
+
"grad_norm": 0.46528419852256775,
|
| 725 |
+
"learning_rate": 0.00029710710304520866,
|
| 726 |
+
"loss": 0.6731,
|
| 727 |
+
"num_input_tokens_seen": 471859200,
|
| 728 |
+
"step": 7200,
|
| 729 |
+
"train_runtime": 4094.7269,
|
| 730 |
+
"train_tokens_per_second": 115235.817
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.073,
|
| 734 |
+
"grad_norm": 0.45732709765434265,
|
| 735 |
+
"learning_rate": 0.00029701332937824885,
|
| 736 |
+
"loss": 0.6667,
|
| 737 |
+
"num_input_tokens_seen": 478412800,
|
| 738 |
+
"step": 7300,
|
| 739 |
+
"train_runtime": 4155.1372,
|
| 740 |
+
"train_tokens_per_second": 115137.666
|
| 741 |
+
},
|
| 742 |
+
{
|
| 743 |
+
"epoch": 0.074,
|
| 744 |
+
"grad_norm": 0.28316208720207214,
|
| 745 |
+
"learning_rate": 0.0002969180752886944,
|
| 746 |
+
"loss": 0.6735,
|
| 747 |
+
"num_input_tokens_seen": 484966400,
|
| 748 |
+
"step": 7400,
|
| 749 |
+
"train_runtime": 4208.4074,
|
| 750 |
+
"train_tokens_per_second": 115237.512
|
| 751 |
+
},
|
| 752 |
+
{
|
| 753 |
+
"epoch": 0.075,
|
| 754 |
+
"grad_norm": 0.35296332836151123,
|
| 755 |
+
"learning_rate": 0.0002968213417357529,
|
| 756 |
+
"loss": 0.674,
|
| 757 |
+
"num_input_tokens_seen": 491520000,
|
| 758 |
+
"step": 7500,
|
| 759 |
+
"train_runtime": 4262.7857,
|
| 760 |
+
"train_tokens_per_second": 115304.882
|
| 761 |
+
},
|
| 762 |
+
{
|
| 763 |
+
"epoch": 0.076,
|
| 764 |
+
"grad_norm": 0.37089261412620544,
|
| 765 |
+
"learning_rate": 0.00029672312969353015,
|
| 766 |
+
"loss": 0.6696,
|
| 767 |
+
"num_input_tokens_seen": 498073600,
|
| 768 |
+
"step": 7600,
|
| 769 |
+
"train_runtime": 4324.696,
|
| 770 |
+
"train_tokens_per_second": 115169.622
|
| 771 |
+
},
|
| 772 |
+
{
|
| 773 |
+
"epoch": 0.077,
|
| 774 |
+
"grad_norm": 0.3451579809188843,
|
| 775 |
+
"learning_rate": 0.00029662344015102027,
|
| 776 |
+
"loss": 0.669,
|
| 777 |
+
"num_input_tokens_seen": 504627200,
|
| 778 |
+
"step": 7700,
|
| 779 |
+
"train_runtime": 4380.0469,
|
| 780 |
+
"train_tokens_per_second": 115210.456
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 0.078,
|
| 784 |
+
"grad_norm": 0.2890200912952423,
|
| 785 |
+
"learning_rate": 0.00029652227411209594,
|
| 786 |
+
"loss": 0.6672,
|
| 787 |
+
"num_input_tokens_seen": 511180800,
|
| 788 |
+
"step": 7800,
|
| 789 |
+
"train_runtime": 4433.6829,
|
| 790 |
+
"train_tokens_per_second": 115294.849
|
| 791 |
+
},
|
| 792 |
+
{
|
| 793 |
+
"epoch": 0.079,
|
| 794 |
+
"grad_norm": 0.3042987287044525,
|
| 795 |
+
"learning_rate": 0.0002964196325954979,
|
| 796 |
+
"loss": 0.6642,
|
| 797 |
+
"num_input_tokens_seen": 517734400,
|
| 798 |
+
"step": 7900,
|
| 799 |
+
"train_runtime": 4488.6662,
|
| 800 |
+
"train_tokens_per_second": 115342.593
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.08,
|
| 804 |
+
"grad_norm": 0.34017419815063477,
|
| 805 |
+
"learning_rate": 0.0002963155166348253,
|
| 806 |
+
"loss": 0.6689,
|
| 807 |
+
"num_input_tokens_seen": 524288000,
|
| 808 |
+
"step": 8000,
|
| 809 |
+
"train_runtime": 4548.362,
|
| 810 |
+
"train_tokens_per_second": 115269.63
|
| 811 |
+
},
|
| 812 |
+
{
|
| 813 |
+
"epoch": 0.081,
|
| 814 |
+
"grad_norm": 0.5584019422531128,
|
| 815 |
+
"learning_rate": 0.0002962099272785246,
|
| 816 |
+
"loss": 0.6703,
|
| 817 |
+
"num_input_tokens_seen": 530841600,
|
| 818 |
+
"step": 8100,
|
| 819 |
+
"train_runtime": 4605.2398,
|
| 820 |
+
"train_tokens_per_second": 115269.047
|
| 821 |
+
},
|
| 822 |
+
{
|
| 823 |
+
"epoch": 0.082,
|
| 824 |
+
"grad_norm": 0.2589603364467621,
|
| 825 |
+
"learning_rate": 0.0002961028655898794,
|
| 826 |
+
"loss": 0.6634,
|
| 827 |
+
"num_input_tokens_seen": 537395200,
|
| 828 |
+
"step": 8200,
|
| 829 |
+
"train_runtime": 4660.2455,
|
| 830 |
+
"train_tokens_per_second": 115314.783
|
| 831 |
+
},
|
| 832 |
+
{
|
| 833 |
+
"epoch": 0.083,
|
| 834 |
+
"grad_norm": 0.41600939631462097,
|
| 835 |
+
"learning_rate": 0.0002959943326469998,
|
| 836 |
+
"loss": 0.6611,
|
| 837 |
+
"num_input_tokens_seen": 543948800,
|
| 838 |
+
"step": 8300,
|
| 839 |
+
"train_runtime": 4714.0847,
|
| 840 |
+
"train_tokens_per_second": 115388.0
|
| 841 |
+
},
|
| 842 |
+
{
|
| 843 |
+
"epoch": 0.084,
|
| 844 |
+
"grad_norm": 0.28658175468444824,
|
| 845 |
+
"learning_rate": 0.0002958843295428112,
|
| 846 |
+
"loss": 0.659,
|
| 847 |
+
"num_input_tokens_seen": 550502400,
|
| 848 |
+
"step": 8400,
|
| 849 |
+
"train_runtime": 4768.8365,
|
| 850 |
+
"train_tokens_per_second": 115437.466
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 0.085,
|
| 854 |
+
"grad_norm": 0.290019154548645,
|
| 855 |
+
"learning_rate": 0.0002957728573850438,
|
| 856 |
+
"loss": 0.661,
|
| 857 |
+
"num_input_tokens_seen": 557056000,
|
| 858 |
+
"step": 8500,
|
| 859 |
+
"train_runtime": 4828.5019,
|
| 860 |
+
"train_tokens_per_second": 115368.289
|
| 861 |
+
},
|
| 862 |
+
{
|
| 863 |
+
"epoch": 0.086,
|
| 864 |
+
"grad_norm": 0.3445529341697693,
|
| 865 |
+
"learning_rate": 0.0002956599172962209,
|
| 866 |
+
"loss": 0.6745,
|
| 867 |
+
"num_input_tokens_seen": 563609600,
|
| 868 |
+
"step": 8600,
|
| 869 |
+
"train_runtime": 4883.6807,
|
| 870 |
+
"train_tokens_per_second": 115406.727
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.087,
|
| 874 |
+
"grad_norm": 0.3387536406517029,
|
| 875 |
+
"learning_rate": 0.0002955455104136479,
|
| 876 |
+
"loss": 0.6596,
|
| 877 |
+
"num_input_tokens_seen": 570163200,
|
| 878 |
+
"step": 8700,
|
| 879 |
+
"train_runtime": 4940.0949,
|
| 880 |
+
"train_tokens_per_second": 115415.435
|
| 881 |
+
},
|
| 882 |
+
{
|
| 883 |
+
"epoch": 0.088,
|
| 884 |
+
"grad_norm": 0.3031691312789917,
|
| 885 |
+
"learning_rate": 0.00029542963788940096,
|
| 886 |
+
"loss": 0.658,
|
| 887 |
+
"num_input_tokens_seen": 576716800,
|
| 888 |
+
"step": 8800,
|
| 889 |
+
"train_runtime": 4993.542,
|
| 890 |
+
"train_tokens_per_second": 115492.529
|
| 891 |
+
},
|
| 892 |
+
{
|
| 893 |
+
"epoch": 0.089,
|
| 894 |
+
"grad_norm": 0.3859446048736572,
|
| 895 |
+
"learning_rate": 0.00029531230089031505,
|
| 896 |
+
"loss": 0.6684,
|
| 897 |
+
"num_input_tokens_seen": 583270400,
|
| 898 |
+
"step": 8900,
|
| 899 |
+
"train_runtime": 5048.437,
|
| 900 |
+
"train_tokens_per_second": 115534.848
|
| 901 |
+
},
|
| 902 |
+
{
|
| 903 |
+
"epoch": 0.09,
|
| 904 |
+
"grad_norm": 0.28773999214172363,
|
| 905 |
+
"learning_rate": 0.0002951935005979724,
|
| 906 |
+
"loss": 0.6585,
|
| 907 |
+
"num_input_tokens_seen": 589824000,
|
| 908 |
+
"step": 9000,
|
| 909 |
+
"train_runtime": 5107.8168,
|
| 910 |
+
"train_tokens_per_second": 115474.775
|
| 911 |
+
},
|
| 912 |
+
{
|
| 913 |
+
"epoch": 0.091,
|
| 914 |
+
"grad_norm": 0.3639370799064636,
|
| 915 |
+
"learning_rate": 0.0002950732382086907,
|
| 916 |
+
"loss": 0.6573,
|
| 917 |
+
"num_input_tokens_seen": 596377600,
|
| 918 |
+
"step": 9100,
|
| 919 |
+
"train_runtime": 5164.5839,
|
| 920 |
+
"train_tokens_per_second": 115474.473
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 0.092,
|
| 924 |
+
"grad_norm": 0.3325030505657196,
|
| 925 |
+
"learning_rate": 0.0002949515149335108,
|
| 926 |
+
"loss": 0.6634,
|
| 927 |
+
"num_input_tokens_seen": 602931200,
|
| 928 |
+
"step": 9200,
|
| 929 |
+
"train_runtime": 5219.0432,
|
| 930 |
+
"train_tokens_per_second": 115525.237
|
| 931 |
+
},
|
| 932 |
+
{
|
| 933 |
+
"epoch": 0.093,
|
| 934 |
+
"grad_norm": 0.3296424448490143,
|
| 935 |
+
"learning_rate": 0.0002948283319981848,
|
| 936 |
+
"loss": 0.65,
|
| 937 |
+
"num_input_tokens_seen": 609484800,
|
| 938 |
+
"step": 9300,
|
| 939 |
+
"train_runtime": 5273.9916,
|
| 940 |
+
"train_tokens_per_second": 115564.234
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.094,
|
| 944 |
+
"grad_norm": 0.323024719953537,
|
| 945 |
+
"learning_rate": 0.00029470369064316354,
|
| 946 |
+
"loss": 0.6533,
|
| 947 |
+
"num_input_tokens_seen": 616038400,
|
| 948 |
+
"step": 9400,
|
| 949 |
+
"train_runtime": 5327.8342,
|
| 950 |
+
"train_tokens_per_second": 115626.421
|
| 951 |
+
},
|
| 952 |
+
{
|
| 953 |
+
"epoch": 0.095,
|
| 954 |
+
"grad_norm": 0.31108859181404114,
|
| 955 |
+
"learning_rate": 0.00029457759212358397,
|
| 956 |
+
"loss": 0.6549,
|
| 957 |
+
"num_input_tokens_seen": 622592000,
|
| 958 |
+
"step": 9500,
|
| 959 |
+
"train_runtime": 5386.4463,
|
| 960 |
+
"train_tokens_per_second": 115584.927
|
| 961 |
+
},
|
| 962 |
+
{
|
| 963 |
+
"epoch": 0.096,
|
| 964 |
+
"grad_norm": 0.43114957213401794,
|
| 965 |
+
"learning_rate": 0.00029445003770925686,
|
| 966 |
+
"loss": 0.6518,
|
| 967 |
+
"num_input_tokens_seen": 629145600,
|
| 968 |
+
"step": 9600,
|
| 969 |
+
"train_runtime": 5441.6464,
|
| 970 |
+
"train_tokens_per_second": 115616.773
|
| 971 |
+
},
|
| 972 |
+
{
|
| 973 |
+
"epoch": 0.097,
|
| 974 |
+
"grad_norm": 0.3269326090812683,
|
| 975 |
+
"learning_rate": 0.00029432102868465367,
|
| 976 |
+
"loss": 0.6539,
|
| 977 |
+
"num_input_tokens_seen": 635699200,
|
| 978 |
+
"step": 9700,
|
| 979 |
+
"train_runtime": 5497.0335,
|
| 980 |
+
"train_tokens_per_second": 115644.046
|
| 981 |
+
},
|
| 982 |
+
{
|
| 983 |
+
"epoch": 0.098,
|
| 984 |
+
"grad_norm": 0.3119284510612488,
|
| 985 |
+
"learning_rate": 0.0002941905663488939,
|
| 986 |
+
"loss": 0.6525,
|
| 987 |
+
"num_input_tokens_seen": 642252800,
|
| 988 |
+
"step": 9800,
|
| 989 |
+
"train_runtime": 5557.4204,
|
| 990 |
+
"train_tokens_per_second": 115566.712
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 0.099,
|
| 994 |
+
"grad_norm": 0.30968910455703735,
|
| 995 |
+
"learning_rate": 0.0002940586520157318,
|
| 996 |
+
"loss": 0.6593,
|
| 997 |
+
"num_input_tokens_seen": 648806400,
|
| 998 |
+
"step": 9900,
|
| 999 |
+
"train_runtime": 5612.1781,
|
| 1000 |
+
"train_tokens_per_second": 115606.881
|
| 1001 |
+
},
|
| 1002 |
+
{
|
| 1003 |
+
"epoch": 0.1,
|
| 1004 |
+
"grad_norm": 0.3167921006679535,
|
| 1005 |
+
"learning_rate": 0.00029392528701354325,
|
| 1006 |
+
"loss": 0.6516,
|
| 1007 |
+
"num_input_tokens_seen": 655360000,
|
| 1008 |
+
"step": 10000,
|
| 1009 |
+
"train_runtime": 5665.7095,
|
| 1010 |
+
"train_tokens_per_second": 115671.303
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.101,
|
| 1014 |
+
"grad_norm": 0.3000660538673401,
|
| 1015 |
+
"learning_rate": 0.00029379047268531243,
|
| 1016 |
+
"loss": 0.6579,
|
| 1017 |
+
"num_input_tokens_seen": 661913600,
|
| 1018 |
+
"step": 10100,
|
| 1019 |
+
"train_runtime": 5721.9414,
|
| 1020 |
+
"train_tokens_per_second": 115679.899
|
| 1021 |
+
},
|
| 1022 |
+
{
|
| 1023 |
+
"epoch": 0.102,
|
| 1024 |
+
"grad_norm": 0.30189189314842224,
|
| 1025 |
+
"learning_rate": 0.00029365421038861795,
|
| 1026 |
+
"loss": 0.6594,
|
| 1027 |
+
"num_input_tokens_seen": 668467200,
|
| 1028 |
+
"step": 10200,
|
| 1029 |
+
"train_runtime": 5777.1263,
|
| 1030 |
+
"train_tokens_per_second": 115709.294
|
| 1031 |
+
},
|
| 1032 |
+
{
|
| 1033 |
+
"epoch": 0.103,
|
| 1034 |
+
"grad_norm": 0.2689358592033386,
|
| 1035 |
+
"learning_rate": 0.0002935165014956198,
|
| 1036 |
+
"loss": 0.6527,
|
| 1037 |
+
"num_input_tokens_seen": 675020800,
|
| 1038 |
+
"step": 10300,
|
| 1039 |
+
"train_runtime": 5832.7222,
|
| 1040 |
+
"train_tokens_per_second": 115729.977
|
| 1041 |
+
},
|
| 1042 |
+
{
|
| 1043 |
+
"epoch": 0.104,
|
| 1044 |
+
"grad_norm": 0.40012845396995544,
|
| 1045 |
+
"learning_rate": 0.0002933773473930448,
|
| 1046 |
+
"loss": 0.6442,
|
| 1047 |
+
"num_input_tokens_seen": 681574400,
|
| 1048 |
+
"step": 10400,
|
| 1049 |
+
"train_runtime": 5888.393,
|
| 1050 |
+
"train_tokens_per_second": 115748.796
|
| 1051 |
+
},
|
| 1052 |
+
{
|
| 1053 |
+
"epoch": 0.105,
|
| 1054 |
+
"grad_norm": 0.26425933837890625,
|
| 1055 |
+
"learning_rate": 0.0002932367494821734,
|
| 1056 |
+
"loss": 0.655,
|
| 1057 |
+
"num_input_tokens_seen": 688128000,
|
| 1058 |
+
"step": 10500,
|
| 1059 |
+
"train_runtime": 5949.719,
|
| 1060 |
+
"train_tokens_per_second": 115657.227
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 0.106,
|
| 1064 |
+
"grad_norm": 0.3352152705192566,
|
| 1065 |
+
"learning_rate": 0.00029309470917882497,
|
| 1066 |
+
"loss": 0.6513,
|
| 1067 |
+
"num_input_tokens_seen": 694681600,
|
| 1068 |
+
"step": 10600,
|
| 1069 |
+
"train_runtime": 6005.1104,
|
| 1070 |
+
"train_tokens_per_second": 115681.738
|
| 1071 |
+
},
|
| 1072 |
+
{
|
| 1073 |
+
"epoch": 0.107,
|
| 1074 |
+
"grad_norm": 0.2809161841869354,
|
| 1075 |
+
"learning_rate": 0.0002929512279133437,
|
| 1076 |
+
"loss": 0.6544,
|
| 1077 |
+
"num_input_tokens_seen": 701235200,
|
| 1078 |
+
"step": 10700,
|
| 1079 |
+
"train_runtime": 6060.3088,
|
| 1080 |
+
"train_tokens_per_second": 115709.482
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 0.108,
|
| 1084 |
+
"grad_norm": 0.29120925068855286,
|
| 1085 |
+
"learning_rate": 0.0002928063071305844,
|
| 1086 |
+
"loss": 0.6502,
|
| 1087 |
+
"num_input_tokens_seen": 707788800,
|
| 1088 |
+
"step": 10800,
|
| 1089 |
+
"train_runtime": 6114.2187,
|
| 1090 |
+
"train_tokens_per_second": 115761.119
|
| 1091 |
+
},
|
| 1092 |
+
{
|
| 1093 |
+
"epoch": 0.109,
|
| 1094 |
+
"grad_norm": 0.312067449092865,
|
| 1095 |
+
"learning_rate": 0.0002926599482898978,
|
| 1096 |
+
"loss": 0.6495,
|
| 1097 |
+
"num_input_tokens_seen": 714342400,
|
| 1098 |
+
"step": 10900,
|
| 1099 |
+
"train_runtime": 6169.1634,
|
| 1100 |
+
"train_tokens_per_second": 115792.426
|
| 1101 |
+
},
|
| 1102 |
+
{
|
| 1103 |
+
"epoch": 0.11,
|
| 1104 |
+
"grad_norm": 0.4553526043891907,
|
| 1105 |
+
"learning_rate": 0.00029251215286511573,
|
| 1106 |
+
"loss": 0.6508,
|
| 1107 |
+
"num_input_tokens_seen": 720896000,
|
| 1108 |
+
"step": 11000,
|
| 1109 |
+
"train_runtime": 6230.1722,
|
| 1110 |
+
"train_tokens_per_second": 115710.445
|
| 1111 |
+
},
|
| 1112 |
+
{
|
| 1113 |
+
"epoch": 0.111,
|
| 1114 |
+
"grad_norm": 0.2880023419857025,
|
| 1115 |
+
"learning_rate": 0.00029236292234453647,
|
| 1116 |
+
"loss": 0.6476,
|
| 1117 |
+
"num_input_tokens_seen": 727449600,
|
| 1118 |
+
"step": 11100,
|
| 1119 |
+
"train_runtime": 6284.648,
|
| 1120 |
+
"train_tokens_per_second": 115750.253
|
| 1121 |
+
},
|
| 1122 |
+
{
|
| 1123 |
+
"epoch": 0.112,
|
| 1124 |
+
"grad_norm": 0.3704841136932373,
|
| 1125 |
+
"learning_rate": 0.0002922122582309097,
|
| 1126 |
+
"loss": 0.6559,
|
| 1127 |
+
"num_input_tokens_seen": 734003200,
|
| 1128 |
+
"step": 11200,
|
| 1129 |
+
"train_runtime": 6340.8669,
|
| 1130 |
+
"train_tokens_per_second": 115757.547
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 0.113,
|
| 1134 |
+
"grad_norm": 0.34455180168151855,
|
| 1135 |
+
"learning_rate": 0.0002920601620414215,
|
| 1136 |
+
"loss": 0.6482,
|
| 1137 |
+
"num_input_tokens_seen": 740556800,
|
| 1138 |
+
"step": 11300,
|
| 1139 |
+
"train_runtime": 6393.5327,
|
| 1140 |
+
"train_tokens_per_second": 115829.046
|
| 1141 |
+
},
|
| 1142 |
+
{
|
| 1143 |
+
"epoch": 0.114,
|
| 1144 |
+
"grad_norm": 0.3323168456554413,
|
| 1145 |
+
"learning_rate": 0.0002919066353076786,
|
| 1146 |
+
"loss": 0.6489,
|
| 1147 |
+
"num_input_tokens_seen": 747110400,
|
| 1148 |
+
"step": 11400,
|
| 1149 |
+
"train_runtime": 6449.7175,
|
| 1150 |
+
"train_tokens_per_second": 115836.144
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.115,
|
| 1154 |
+
"grad_norm": 0.4480803608894348,
|
| 1155 |
+
"learning_rate": 0.00029175167957569366,
|
| 1156 |
+
"loss": 0.6492,
|
| 1157 |
+
"num_input_tokens_seen": 753664000,
|
| 1158 |
+
"step": 11500,
|
| 1159 |
+
"train_runtime": 6503.6523,
|
| 1160 |
+
"train_tokens_per_second": 115883.194
|
| 1161 |
+
},
|
| 1162 |
+
{
|
| 1163 |
+
"epoch": 0.116,
|
| 1164 |
+
"grad_norm": 0.2677996754646301,
|
| 1165 |
+
"learning_rate": 0.0002915952964058691,
|
| 1166 |
+
"loss": 0.6462,
|
| 1167 |
+
"num_input_tokens_seen": 760217600,
|
| 1168 |
+
"step": 11600,
|
| 1169 |
+
"train_runtime": 6565.0474,
|
| 1170 |
+
"train_tokens_per_second": 115797.731
|
| 1171 |
+
},
|
| 1172 |
+
{
|
| 1173 |
+
"epoch": 0.117,
|
| 1174 |
+
"grad_norm": 0.2838340699672699,
|
| 1175 |
+
"learning_rate": 0.00029143748737298173,
|
| 1176 |
+
"loss": 0.6574,
|
| 1177 |
+
"num_input_tokens_seen": 766771200,
|
| 1178 |
+
"step": 11700,
|
| 1179 |
+
"train_runtime": 6620.4342,
|
| 1180 |
+
"train_tokens_per_second": 115818.868
|
| 1181 |
+
},
|
| 1182 |
+
{
|
| 1183 |
+
"epoch": 0.118,
|
| 1184 |
+
"grad_norm": 0.3221312165260315,
|
| 1185 |
+
"learning_rate": 0.00029127825406616677,
|
| 1186 |
+
"loss": 0.6456,
|
| 1187 |
+
"num_input_tokens_seen": 773324800,
|
| 1188 |
+
"step": 11800,
|
| 1189 |
+
"train_runtime": 6676.1674,
|
| 1190 |
+
"train_tokens_per_second": 115833.645
|
| 1191 |
+
},
|
| 1192 |
+
{
|
| 1193 |
+
"epoch": 0.119,
|
| 1194 |
+
"grad_norm": 0.2825019657611847,
|
| 1195 |
+
"learning_rate": 0.0002911175980889019,
|
| 1196 |
+
"loss": 0.6526,
|
| 1197 |
+
"num_input_tokens_seen": 779878400,
|
| 1198 |
+
"step": 11900,
|
| 1199 |
+
"train_runtime": 6730.5574,
|
| 1200 |
+
"train_tokens_per_second": 115871.295
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 0.12,
|
| 1204 |
+
"grad_norm": 0.38775157928466797,
|
| 1205 |
+
"learning_rate": 0.00029095552105899095,
|
| 1206 |
+
"loss": 0.6454,
|
| 1207 |
+
"num_input_tokens_seen": 786432000,
|
| 1208 |
+
"step": 12000,
|
| 1209 |
+
"train_runtime": 6784.0533,
|
| 1210 |
+
"train_tokens_per_second": 115923.617
|
| 1211 |
+
},
|
| 1212 |
+
{
|
| 1213 |
+
"epoch": 0.121,
|
| 1214 |
+
"grad_norm": 0.32168275117874146,
|
| 1215 |
+
"learning_rate": 0.0002907920246085478,
|
| 1216 |
+
"loss": 0.644,
|
| 1217 |
+
"num_input_tokens_seen": 792985600,
|
| 1218 |
+
"step": 12100,
|
| 1219 |
+
"train_runtime": 6844.3129,
|
| 1220 |
+
"train_tokens_per_second": 115860.512
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 0.122,
|
| 1224 |
+
"grad_norm": 0.3085991442203522,
|
| 1225 |
+
"learning_rate": 0.00029062711038397996,
|
| 1226 |
+
"loss": 0.6609,
|
| 1227 |
+
"num_input_tokens_seen": 799539200,
|
| 1228 |
+
"step": 12200,
|
| 1229 |
+
"train_runtime": 6899.148,
|
| 1230 |
+
"train_tokens_per_second": 115889.557
|
| 1231 |
+
},
|
| 1232 |
+
{
|
| 1233 |
+
"epoch": 0.123,
|
| 1234 |
+
"grad_norm": 0.26165974140167236,
|
| 1235 |
+
"learning_rate": 0.00029046078004597175,
|
| 1236 |
+
"loss": 0.6436,
|
| 1237 |
+
"num_input_tokens_seen": 806092800,
|
| 1238 |
+
"step": 12300,
|
| 1239 |
+
"train_runtime": 6953.2129,
|
| 1240 |
+
"train_tokens_per_second": 115930.983
|
| 1241 |
+
},
|
| 1242 |
+
{
|
| 1243 |
+
"epoch": 0.124,
|
| 1244 |
+
"grad_norm": 0.5689886212348938,
|
| 1245 |
+
"learning_rate": 0.00029029303526946796,
|
| 1246 |
+
"loss": 0.6432,
|
| 1247 |
+
"num_input_tokens_seen": 812646400,
|
| 1248 |
+
"step": 12400,
|
| 1249 |
+
"train_runtime": 7008.0879,
|
| 1250 |
+
"train_tokens_per_second": 115958.363
|
| 1251 |
+
},
|
| 1252 |
+
{
|
| 1253 |
+
"epoch": 0.125,
|
| 1254 |
+
"grad_norm": 0.3152099847793579,
|
| 1255 |
+
"learning_rate": 0.0002901238777436565,
|
| 1256 |
+
"loss": 0.6391,
|
| 1257 |
+
"num_input_tokens_seen": 819200000,
|
| 1258 |
+
"step": 12500,
|
| 1259 |
+
"train_runtime": 7067.2869,
|
| 1260 |
+
"train_tokens_per_second": 115914.355
|
| 1261 |
+
},
|
| 1262 |
+
{
|
| 1263 |
+
"epoch": 0.126,
|
| 1264 |
+
"grad_norm": 0.28095683455467224,
|
| 1265 |
+
"learning_rate": 0.00028995330917195184,
|
| 1266 |
+
"loss": 0.645,
|
| 1267 |
+
"num_input_tokens_seen": 825753600,
|
| 1268 |
+
"step": 12600,
|
| 1269 |
+
"train_runtime": 7122.3035,
|
| 1270 |
+
"train_tokens_per_second": 115939.121
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 0.127,
|
| 1274 |
+
"grad_norm": 0.271088570356369,
|
| 1275 |
+
"learning_rate": 0.00028978133127197765,
|
| 1276 |
+
"loss": 0.6452,
|
| 1277 |
+
"num_input_tokens_seen": 832307200,
|
| 1278 |
+
"step": 12700,
|
| 1279 |
+
"train_runtime": 7177.0079,
|
| 1280 |
+
"train_tokens_per_second": 115968.55
|
| 1281 |
+
},
|
| 1282 |
+
{
|
| 1283 |
+
"epoch": 0.128,
|
| 1284 |
+
"grad_norm": 0.3357028067111969,
|
| 1285 |
+
"learning_rate": 0.0002896079457755493,
|
| 1286 |
+
"loss": 0.6477,
|
| 1287 |
+
"num_input_tokens_seen": 838860800,
|
| 1288 |
+
"step": 12800,
|
| 1289 |
+
"train_runtime": 7232.8071,
|
| 1290 |
+
"train_tokens_per_second": 115979.977
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 0.129,
|
| 1294 |
+
"grad_norm": 0.4945499897003174,
|
| 1295 |
+
"learning_rate": 0.000289433154428657,
|
| 1296 |
+
"loss": 0.6455,
|
| 1297 |
+
"num_input_tokens_seen": 845414400,
|
| 1298 |
+
"step": 12900,
|
| 1299 |
+
"train_runtime": 7287.7238,
|
| 1300 |
+
"train_tokens_per_second": 116005.274
|
| 1301 |
+
},
|
| 1302 |
+
{
|
| 1303 |
+
"epoch": 0.13,
|
| 1304 |
+
"grad_norm": 0.43597710132598877,
|
| 1305 |
+
"learning_rate": 0.0002892569589914476,
|
| 1306 |
+
"loss": 0.6422,
|
| 1307 |
+
"num_input_tokens_seen": 851968000,
|
| 1308 |
+
"step": 13000,
|
| 1309 |
+
"train_runtime": 7347.1365,
|
| 1310 |
+
"train_tokens_per_second": 115959.19
|
| 1311 |
+
},
|
| 1312 |
+
{
|
| 1313 |
+
"epoch": 0.131,
|
| 1314 |
+
"grad_norm": 0.2706185281276703,
|
| 1315 |
+
"learning_rate": 0.0002890793612382072,
|
| 1316 |
+
"loss": 0.6434,
|
| 1317 |
+
"num_input_tokens_seen": 858521600,
|
| 1318 |
+
"step": 13100,
|
| 1319 |
+
"train_runtime": 7401.7618,
|
| 1320 |
+
"train_tokens_per_second": 115988.818
|
| 1321 |
+
},
|
| 1322 |
+
{
|
| 1323 |
+
"epoch": 0.132,
|
| 1324 |
+
"grad_norm": 0.22793404757976532,
|
| 1325 |
+
"learning_rate": 0.0002889003629573432,
|
| 1326 |
+
"loss": 0.646,
|
| 1327 |
+
"num_input_tokens_seen": 865075200,
|
| 1328 |
+
"step": 13200,
|
| 1329 |
+
"train_runtime": 7457.327,
|
| 1330 |
+
"train_tokens_per_second": 116003.389
|
| 1331 |
+
},
|
| 1332 |
+
{
|
| 1333 |
+
"epoch": 0.133,
|
| 1334 |
+
"grad_norm": 0.2563342750072479,
|
| 1335 |
+
"learning_rate": 0.00028871996595136626,
|
| 1336 |
+
"loss": 0.6503,
|
| 1337 |
+
"num_input_tokens_seen": 871628800,
|
| 1338 |
+
"step": 13300,
|
| 1339 |
+
"train_runtime": 7511.6003,
|
| 1340 |
+
"train_tokens_per_second": 116037.697
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 0.134,
|
| 1344 |
+
"grad_norm": 0.30800577998161316,
|
| 1345 |
+
"learning_rate": 0.0002885381720368723,
|
| 1346 |
+
"loss": 0.6378,
|
| 1347 |
+
"num_input_tokens_seen": 878182400,
|
| 1348 |
+
"step": 13400,
|
| 1349 |
+
"train_runtime": 7571.585,
|
| 1350 |
+
"train_tokens_per_second": 115983.959
|
| 1351 |
+
},
|
| 1352 |
+
{
|
| 1353 |
+
"epoch": 0.135,
|
| 1354 |
+
"grad_norm": 0.2970588505268097,
|
| 1355 |
+
"learning_rate": 0.000288354983044524,
|
| 1356 |
+
"loss": 0.6375,
|
| 1357 |
+
"num_input_tokens_seen": 884736000,
|
| 1358 |
+
"step": 13500,
|
| 1359 |
+
"train_runtime": 7625.682,
|
| 1360 |
+
"train_tokens_per_second": 116020.573
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 0.136,
|
| 1364 |
+
"grad_norm": 0.24373579025268555,
|
| 1365 |
+
"learning_rate": 0.00028817040081903245,
|
| 1366 |
+
"loss": 0.6436,
|
| 1367 |
+
"num_input_tokens_seen": 891289600,
|
| 1368 |
+
"step": 13600,
|
| 1369 |
+
"train_runtime": 7679.3213,
|
| 1370 |
+
"train_tokens_per_second": 116063.591
|
| 1371 |
+
},
|
| 1372 |
+
{
|
| 1373 |
+
"epoch": 0.137,
|
| 1374 |
+
"grad_norm": 0.6528549194335938,
|
| 1375 |
+
"learning_rate": 0.00028798442721913867,
|
| 1376 |
+
"loss": 0.639,
|
| 1377 |
+
"num_input_tokens_seen": 897843200,
|
| 1378 |
+
"step": 13700,
|
| 1379 |
+
"train_runtime": 7735.7681,
|
| 1380 |
+
"train_tokens_per_second": 116063.872
|
| 1381 |
+
},
|
| 1382 |
+
{
|
| 1383 |
+
"epoch": 0.138,
|
| 1384 |
+
"grad_norm": 0.2750374376773834,
|
| 1385 |
+
"learning_rate": 0.00028779706411759465,
|
| 1386 |
+
"loss": 0.6362,
|
| 1387 |
+
"num_input_tokens_seen": 904396800,
|
| 1388 |
+
"step": 13800,
|
| 1389 |
+
"train_runtime": 7796.3703,
|
| 1390 |
+
"train_tokens_per_second": 116002.288
|
| 1391 |
+
},
|
| 1392 |
+
{
|
| 1393 |
+
"epoch": 0.139,
|
| 1394 |
+
"grad_norm": 0.3658313751220703,
|
| 1395 |
+
"learning_rate": 0.00028760831340114484,
|
| 1396 |
+
"loss": 0.6424,
|
| 1397 |
+
"num_input_tokens_seen": 910950400,
|
| 1398 |
+
"step": 13900,
|
| 1399 |
+
"train_runtime": 7851.8089,
|
| 1400 |
+
"train_tokens_per_second": 116017.903
|
| 1401 |
+
},
|
| 1402 |
+
{
|
| 1403 |
+
"epoch": 0.14,
|
| 1404 |
+
"grad_norm": 0.2838168740272522,
|
| 1405 |
+
"learning_rate": 0.00028741817697050683,
|
| 1406 |
+
"loss": 0.6424,
|
| 1407 |
+
"num_input_tokens_seen": 917504000,
|
| 1408 |
+
"step": 14000,
|
| 1409 |
+
"train_runtime": 7906.2377,
|
| 1410 |
+
"train_tokens_per_second": 116048.117
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 0.141,
|
| 1414 |
+
"grad_norm": 0.30455613136291504,
|
| 1415 |
+
"learning_rate": 0.00028722665674035233,
|
| 1416 |
+
"loss": 0.637,
|
| 1417 |
+
"num_input_tokens_seen": 924057600,
|
| 1418 |
+
"step": 14100,
|
| 1419 |
+
"train_runtime": 7961.9367,
|
| 1420 |
+
"train_tokens_per_second": 116059.401
|
| 1421 |
+
},
|
| 1422 |
+
{
|
| 1423 |
+
"epoch": 0.142,
|
| 1424 |
+
"grad_norm": 0.3287622630596161,
|
| 1425 |
+
"learning_rate": 0.0002870337546392879,
|
| 1426 |
+
"loss": 0.6612,
|
| 1427 |
+
"num_input_tokens_seen": 930611200,
|
| 1428 |
+
"step": 14200,
|
| 1429 |
+
"train_runtime": 8016.7903,
|
| 1430 |
+
"train_tokens_per_second": 116082.767
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 0.143,
|
| 1434 |
+
"grad_norm": 0.2602579593658447,
|
| 1435 |
+
"learning_rate": 0.00028683947260983576,
|
| 1436 |
+
"loss": 0.6433,
|
| 1437 |
+
"num_input_tokens_seen": 937164800,
|
| 1438 |
+
"step": 14300,
|
| 1439 |
+
"train_runtime": 8072.3352,
|
| 1440 |
+
"train_tokens_per_second": 116095.873
|
| 1441 |
+
},
|
| 1442 |
+
{
|
| 1443 |
+
"epoch": 0.144,
|
| 1444 |
+
"grad_norm": 0.4514355957508087,
|
| 1445 |
+
"learning_rate": 0.00028664381260841356,
|
| 1446 |
+
"loss": 0.6346,
|
| 1447 |
+
"num_input_tokens_seen": 943718400,
|
| 1448 |
+
"step": 14400,
|
| 1449 |
+
"train_runtime": 8133.1299,
|
| 1450 |
+
"train_tokens_per_second": 116033.853
|
| 1451 |
+
},
|
| 1452 |
+
{
|
| 1453 |
+
"epoch": 0.145,
|
| 1454 |
+
"grad_norm": 0.7613756060600281,
|
| 1455 |
+
"learning_rate": 0.0002864467766053154,
|
| 1456 |
+
"loss": 0.6383,
|
| 1457 |
+
"num_input_tokens_seen": 950272000,
|
| 1458 |
+
"step": 14500,
|
| 1459 |
+
"train_runtime": 8186.629,
|
| 1460 |
+
"train_tokens_per_second": 116076.103
|
| 1461 |
+
},
|
| 1462 |
+
{
|
| 1463 |
+
"epoch": 0.146,
|
| 1464 |
+
"grad_norm": 0.28044015169143677,
|
| 1465 |
+
"learning_rate": 0.00028624836658469165,
|
| 1466 |
+
"loss": 0.6363,
|
| 1467 |
+
"num_input_tokens_seen": 956825600,
|
| 1468 |
+
"step": 14600,
|
| 1469 |
+
"train_runtime": 8246.5064,
|
| 1470 |
+
"train_tokens_per_second": 116027.995
|
| 1471 |
+
},
|
| 1472 |
+
{
|
| 1473 |
+
"epoch": 0.147,
|
| 1474 |
+
"grad_norm": 0.3139192759990692,
|
| 1475 |
+
"learning_rate": 0.00028604858454452906,
|
| 1476 |
+
"loss": 0.65,
|
| 1477 |
+
"num_input_tokens_seen": 963379200,
|
| 1478 |
+
"step": 14700,
|
| 1479 |
+
"train_runtime": 8296.4517,
|
| 1480 |
+
"train_tokens_per_second": 116119.424
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"epoch": 0.148,
|
| 1484 |
+
"grad_norm": 0.33300819993019104,
|
| 1485 |
+
"learning_rate": 0.00028584743249663057,
|
| 1486 |
+
"loss": 0.6409,
|
| 1487 |
+
"num_input_tokens_seen": 969932800,
|
| 1488 |
+
"step": 14800,
|
| 1489 |
+
"train_runtime": 8351.7667,
|
| 1490 |
+
"train_tokens_per_second": 116135.045
|
| 1491 |
+
},
|
| 1492 |
+
{
|
| 1493 |
+
"epoch": 0.149,
|
| 1494 |
+
"grad_norm": 0.35637542605400085,
|
| 1495 |
+
"learning_rate": 0.000285644912466595,
|
| 1496 |
+
"loss": 0.6356,
|
| 1497 |
+
"num_input_tokens_seen": 976486400,
|
| 1498 |
+
"step": 14900,
|
| 1499 |
+
"train_runtime": 8412.3756,
|
| 1500 |
+
"train_tokens_per_second": 116077.366
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"epoch": 0.15,
|
| 1504 |
+
"grad_norm": 0.2518846392631531,
|
| 1505 |
+
"learning_rate": 0.00028544102649379684,
|
| 1506 |
+
"loss": 0.6435,
|
| 1507 |
+
"num_input_tokens_seen": 983040000,
|
| 1508 |
+
"step": 15000,
|
| 1509 |
+
"train_runtime": 8466.8259,
|
| 1510 |
+
"train_tokens_per_second": 116104.904
|
| 1511 |
+
},
|
| 1512 |
+
{
|
| 1513 |
+
"epoch": 0.151,
|
| 1514 |
+
"grad_norm": 0.2983591854572296,
|
| 1515 |
+
"learning_rate": 0.00028523577663136556,
|
| 1516 |
+
"loss": 0.638,
|
| 1517 |
+
"num_input_tokens_seen": 989593600,
|
| 1518 |
+
"step": 15100,
|
| 1519 |
+
"train_runtime": 8526.2848,
|
| 1520 |
+
"train_tokens_per_second": 116063.869
|
| 1521 |
+
},
|
| 1522 |
+
{
|
| 1523 |
+
"epoch": 0.152,
|
| 1524 |
+
"grad_norm": 0.26403748989105225,
|
| 1525 |
+
"learning_rate": 0.000285029164946165,
|
| 1526 |
+
"loss": 0.6439,
|
| 1527 |
+
"num_input_tokens_seen": 996147200,
|
| 1528 |
+
"step": 15200,
|
| 1529 |
+
"train_runtime": 8581.358,
|
| 1530 |
+
"train_tokens_per_second": 116082.7
|
| 1531 |
+
},
|
| 1532 |
+
{
|
| 1533 |
+
"epoch": 0.153,
|
| 1534 |
+
"grad_norm": 0.3652186393737793,
|
| 1535 |
+
"learning_rate": 0.0002848211935187725,
|
| 1536 |
+
"loss": 0.6499,
|
| 1537 |
+
"num_input_tokens_seen": 1002700800,
|
| 1538 |
+
"step": 15300,
|
| 1539 |
+
"train_runtime": 8637.0354,
|
| 1540 |
+
"train_tokens_per_second": 116093.167
|
| 1541 |
+
},
|
| 1542 |
+
{
|
| 1543 |
+
"epoch": 0.154,
|
| 1544 |
+
"grad_norm": 0.3816509246826172,
|
| 1545 |
+
"learning_rate": 0.0002846118644434581,
|
| 1546 |
+
"loss": 0.6355,
|
| 1547 |
+
"num_input_tokens_seen": 1009254400,
|
| 1548 |
+
"step": 15400,
|
| 1549 |
+
"train_runtime": 8691.2267,
|
| 1550 |
+
"train_tokens_per_second": 116123.355
|
| 1551 |
+
},
|
| 1552 |
+
{
|
| 1553 |
+
"epoch": 0.155,
|
| 1554 |
+
"grad_norm": 0.35392388701438904,
|
| 1555 |
+
"learning_rate": 0.00028440117982816326,
|
| 1556 |
+
"loss": 0.6286,
|
| 1557 |
+
"num_input_tokens_seen": 1015808000,
|
| 1558 |
+
"step": 15500,
|
| 1559 |
+
"train_runtime": 8745.8114,
|
| 1560 |
+
"train_tokens_per_second": 116147.943
|
| 1561 |
+
},
|
| 1562 |
+
{
|
| 1563 |
+
"epoch": 0.156,
|
| 1564 |
+
"grad_norm": 0.34099990129470825,
|
| 1565 |
+
"learning_rate": 0.0002841891417944796,
|
| 1566 |
+
"loss": 0.6396,
|
| 1567 |
+
"num_input_tokens_seen": 1022361600,
|
| 1568 |
+
"step": 15600,
|
| 1569 |
+
"train_runtime": 8805.4075,
|
| 1570 |
+
"train_tokens_per_second": 116106.108
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 0.157,
|
| 1574 |
+
"grad_norm": 0.2872321605682373,
|
| 1575 |
+
"learning_rate": 0.0002839757524776279,
|
| 1576 |
+
"loss": 0.6436,
|
| 1577 |
+
"num_input_tokens_seen": 1028915200,
|
| 1578 |
+
"step": 15700,
|
| 1579 |
+
"train_runtime": 8860.6468,
|
| 1580 |
+
"train_tokens_per_second": 116121.907
|
| 1581 |
+
},
|
| 1582 |
+
{
|
| 1583 |
+
"epoch": 0.158,
|
| 1584 |
+
"grad_norm": 0.2647675573825836,
|
| 1585 |
+
"learning_rate": 0.0002837610140264361,
|
| 1586 |
+
"loss": 0.636,
|
| 1587 |
+
"num_input_tokens_seen": 1035468800,
|
| 1588 |
+
"step": 15800,
|
| 1589 |
+
"train_runtime": 8915.1622,
|
| 1590 |
+
"train_tokens_per_second": 116146.94
|
| 1591 |
+
},
|
| 1592 |
+
{
|
| 1593 |
+
"epoch": 0.159,
|
| 1594 |
+
"grad_norm": 0.28176337480545044,
|
| 1595 |
+
"learning_rate": 0.0002835449286033182,
|
| 1596 |
+
"loss": 0.6322,
|
| 1597 |
+
"num_input_tokens_seen": 1042022400,
|
| 1598 |
+
"step": 15900,
|
| 1599 |
+
"train_runtime": 8975.5504,
|
| 1600 |
+
"train_tokens_per_second": 116095.655
|
| 1601 |
+
},
|
| 1602 |
+
{
|
| 1603 |
+
"epoch": 0.16,
|
| 1604 |
+
"grad_norm": 0.31414464116096497,
|
| 1605 |
+
"learning_rate": 0.0002833274983842518,
|
| 1606 |
+
"loss": 0.6282,
|
| 1607 |
+
"num_input_tokens_seen": 1048576000,
|
| 1608 |
+
"step": 16000,
|
| 1609 |
+
"train_runtime": 9029.8157,
|
| 1610 |
+
"train_tokens_per_second": 116123.744
|
| 1611 |
+
},
|
| 1612 |
+
{
|
| 1613 |
+
"epoch": 0.161,
|
| 1614 |
+
"grad_norm": 0.3065328598022461,
|
| 1615 |
+
"learning_rate": 0.0002831087255587569,
|
| 1616 |
+
"loss": 0.6333,
|
| 1617 |
+
"num_input_tokens_seen": 1055129600,
|
| 1618 |
+
"step": 16100,
|
| 1619 |
+
"train_runtime": 9084.3619,
|
| 1620 |
+
"train_tokens_per_second": 116147.905
|
| 1621 |
+
},
|
| 1622 |
+
{
|
| 1623 |
+
"epoch": 0.162,
|
| 1624 |
+
"grad_norm": 0.28930503129959106,
|
| 1625 |
+
"learning_rate": 0.0002828886123298734,
|
| 1626 |
+
"loss": 0.6333,
|
| 1627 |
+
"num_input_tokens_seen": 1061683200,
|
| 1628 |
+
"step": 16200,
|
| 1629 |
+
"train_runtime": 9139.4895,
|
| 1630 |
+
"train_tokens_per_second": 116164.388
|
| 1631 |
+
},
|
| 1632 |
+
{
|
| 1633 |
+
"epoch": 0.163,
|
| 1634 |
+
"grad_norm": 0.28349098563194275,
|
| 1635 |
+
"learning_rate": 0.00028266716091413906,
|
| 1636 |
+
"loss": 0.6368,
|
| 1637 |
+
"num_input_tokens_seen": 1068236800,
|
| 1638 |
+
"step": 16300,
|
| 1639 |
+
"train_runtime": 9194.7322,
|
| 1640 |
+
"train_tokens_per_second": 116179.219
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 0.164,
|
| 1644 |
+
"grad_norm": 0.2510078251361847,
|
| 1645 |
+
"learning_rate": 0.0002824443735415673,
|
| 1646 |
+
"loss": 0.6348,
|
| 1647 |
+
"num_input_tokens_seen": 1074790400,
|
| 1648 |
+
"step": 16400,
|
| 1649 |
+
"train_runtime": 9254.9047,
|
| 1650 |
+
"train_tokens_per_second": 116131.979
|
| 1651 |
+
},
|
| 1652 |
+
{
|
| 1653 |
+
"epoch": 0.165,
|
| 1654 |
+
"grad_norm": 0.35605230927467346,
|
| 1655 |
+
"learning_rate": 0.0002822202524556243,
|
| 1656 |
+
"loss": 0.639,
|
| 1657 |
+
"num_input_tokens_seen": 1081344000,
|
| 1658 |
+
"step": 16500,
|
| 1659 |
+
"train_runtime": 9310.1385,
|
| 1660 |
+
"train_tokens_per_second": 116146.929
|
| 1661 |
+
},
|
| 1662 |
+
{
|
| 1663 |
+
"epoch": 0.166,
|
| 1664 |
+
"grad_norm": 0.518841564655304,
|
| 1665 |
+
"learning_rate": 0.00028199479991320695,
|
| 1666 |
+
"loss": 0.6424,
|
| 1667 |
+
"num_input_tokens_seen": 1087897600,
|
| 1668 |
+
"step": 16600,
|
| 1669 |
+
"train_runtime": 9365.0545,
|
| 1670 |
+
"train_tokens_per_second": 116165.645
|
| 1671 |
+
},
|
| 1672 |
+
{
|
| 1673 |
+
"epoch": 0.167,
|
| 1674 |
+
"grad_norm": 0.2293197512626648,
|
| 1675 |
+
"learning_rate": 0.00028176801818461994,
|
| 1676 |
+
"loss": 0.6316,
|
| 1677 |
+
"num_input_tokens_seen": 1094451200,
|
| 1678 |
+
"step": 16700,
|
| 1679 |
+
"train_runtime": 9419.4884,
|
| 1680 |
+
"train_tokens_per_second": 116190.09
|
| 1681 |
+
},
|
| 1682 |
+
{
|
| 1683 |
+
"epoch": 0.168,
|
| 1684 |
+
"grad_norm": 0.2790850102901459,
|
| 1685 |
+
"learning_rate": 0.00028153990955355273,
|
| 1686 |
+
"loss": 0.636,
|
| 1687 |
+
"num_input_tokens_seen": 1101004800,
|
| 1688 |
+
"step": 16800,
|
| 1689 |
+
"train_runtime": 9476.046,
|
| 1690 |
+
"train_tokens_per_second": 116188.207
|
| 1691 |
+
},
|
| 1692 |
+
{
|
| 1693 |
+
"epoch": 0.169,
|
| 1694 |
+
"grad_norm": 0.2565983235836029,
|
| 1695 |
+
"learning_rate": 0.00028131047631705665,
|
| 1696 |
+
"loss": 0.6351,
|
| 1697 |
+
"num_input_tokens_seen": 1107558400,
|
| 1698 |
+
"step": 16900,
|
| 1699 |
+
"train_runtime": 9536.06,
|
| 1700 |
+
"train_tokens_per_second": 116144.236
|
| 1701 |
+
},
|
| 1702 |
+
{
|
| 1703 |
+
"epoch": 0.17,
|
| 1704 |
+
"grad_norm": 0.2592833638191223,
|
| 1705 |
+
"learning_rate": 0.00028107972078552187,
|
| 1706 |
+
"loss": 0.637,
|
| 1707 |
+
"num_input_tokens_seen": 1114112000,
|
| 1708 |
+
"step": 17000,
|
| 1709 |
+
"train_runtime": 9591.4651,
|
| 1710 |
+
"train_tokens_per_second": 116156.603
|
| 1711 |
+
},
|
| 1712 |
+
{
|
| 1713 |
+
"epoch": 0.171,
|
| 1714 |
+
"grad_norm": 0.2744060754776001,
|
| 1715 |
+
"learning_rate": 0.0002808476452826541,
|
| 1716 |
+
"loss": 0.6313,
|
| 1717 |
+
"num_input_tokens_seen": 1120665600,
|
| 1718 |
+
"step": 17100,
|
| 1719 |
+
"train_runtime": 9644.9448,
|
| 1720 |
+
"train_tokens_per_second": 116192.018
|
| 1721 |
+
},
|
| 1722 |
+
{
|
| 1723 |
+
"epoch": 0.172,
|
| 1724 |
+
"grad_norm": 0.2991725504398346,
|
| 1725 |
+
"learning_rate": 0.00028061425214545094,
|
| 1726 |
+
"loss": 0.6298,
|
| 1727 |
+
"num_input_tokens_seen": 1127219200,
|
| 1728 |
+
"step": 17200,
|
| 1729 |
+
"train_runtime": 9706.1229,
|
| 1730 |
+
"train_tokens_per_second": 116134.857
|
| 1731 |
+
},
|
| 1732 |
+
{
|
| 1733 |
+
"epoch": 0.173,
|
| 1734 |
+
"grad_norm": 0.3391658365726471,
|
| 1735 |
+
"learning_rate": 0.00028037954372417883,
|
| 1736 |
+
"loss": 0.6316,
|
| 1737 |
+
"num_input_tokens_seen": 1133772800,
|
| 1738 |
+
"step": 17300,
|
| 1739 |
+
"train_runtime": 9760.541,
|
| 1740 |
+
"train_tokens_per_second": 116158.807
|
| 1741 |
+
},
|
| 1742 |
+
{
|
| 1743 |
+
"epoch": 0.174,
|
| 1744 |
+
"grad_norm": 0.3975388705730438,
|
| 1745 |
+
"learning_rate": 0.0002801435223823488,
|
| 1746 |
+
"loss": 0.639,
|
| 1747 |
+
"num_input_tokens_seen": 1140326400,
|
| 1748 |
+
"step": 17400,
|
| 1749 |
+
"train_runtime": 9816.1595,
|
| 1750 |
+
"train_tokens_per_second": 116168.284
|
| 1751 |
+
},
|
| 1752 |
+
{
|
| 1753 |
+
"epoch": 0.175,
|
| 1754 |
+
"grad_norm": 0.3754902482032776,
|
| 1755 |
+
"learning_rate": 0.00027990619049669336,
|
| 1756 |
+
"loss": 0.6391,
|
| 1757 |
+
"num_input_tokens_seen": 1146880000,
|
| 1758 |
+
"step": 17500,
|
| 1759 |
+
"train_runtime": 9870.38,
|
| 1760 |
+
"train_tokens_per_second": 116194.108
|
| 1761 |
+
},
|
| 1762 |
+
{
|
| 1763 |
+
"epoch": 0.176,
|
| 1764 |
+
"grad_norm": 0.2742336094379425,
|
| 1765 |
+
"learning_rate": 0.00027966755045714177,
|
| 1766 |
+
"loss": 0.6425,
|
| 1767 |
+
"num_input_tokens_seen": 1153433600,
|
| 1768 |
+
"step": 17600,
|
| 1769 |
+
"train_runtime": 9924.9127,
|
| 1770 |
+
"train_tokens_per_second": 116215.994
|
| 1771 |
+
},
|
| 1772 |
+
{
|
| 1773 |
+
"epoch": 0.177,
|
| 1774 |
+
"grad_norm": 0.31107792258262634,
|
| 1775 |
+
"learning_rate": 0.00027942760466679673,
|
| 1776 |
+
"loss": 0.6312,
|
| 1777 |
+
"num_input_tokens_seen": 1159987200,
|
| 1778 |
+
"step": 17700,
|
| 1779 |
+
"train_runtime": 9985.7287,
|
| 1780 |
+
"train_tokens_per_second": 116164.502
|
| 1781 |
+
},
|
| 1782 |
+
{
|
| 1783 |
+
"epoch": 0.178,
|
| 1784 |
+
"grad_norm": 0.2781549096107483,
|
| 1785 |
+
"learning_rate": 0.00027918635554190956,
|
| 1786 |
+
"loss": 0.6447,
|
| 1787 |
+
"num_input_tokens_seen": 1166540800,
|
| 1788 |
+
"step": 17800,
|
| 1789 |
+
"train_runtime": 10039.8303,
|
| 1790 |
+
"train_tokens_per_second": 116191.287
|
| 1791 |
+
},
|
| 1792 |
+
{
|
| 1793 |
+
"epoch": 0.179,
|
| 1794 |
+
"grad_norm": 0.28636643290519714,
|
| 1795 |
+
"learning_rate": 0.00027894380551185636,
|
| 1796 |
+
"loss": 0.6383,
|
| 1797 |
+
"num_input_tokens_seen": 1173094400,
|
| 1798 |
+
"step": 17900,
|
| 1799 |
+
"train_runtime": 10096.4761,
|
| 1800 |
+
"train_tokens_per_second": 116188.498
|
| 1801 |
+
},
|
| 1802 |
+
{
|
| 1803 |
+
"epoch": 0.18,
|
| 1804 |
+
"grad_norm": 0.25008583068847656,
|
| 1805 |
+
"learning_rate": 0.00027869995701911314,
|
| 1806 |
+
"loss": 0.6286,
|
| 1807 |
+
"num_input_tokens_seen": 1179648000,
|
| 1808 |
+
"step": 18000,
|
| 1809 |
+
"train_runtime": 10150.5104,
|
| 1810 |
+
"train_tokens_per_second": 116215.634
|
| 1811 |
+
},
|
| 1812 |
+
{
|
| 1813 |
+
"epoch": 0.181,
|
| 1814 |
+
"grad_norm": 0.2600831985473633,
|
| 1815 |
+
"learning_rate": 0.0002784548125192316,
|
| 1816 |
+
"loss": 0.6265,
|
| 1817 |
+
"num_input_tokens_seen": 1186201600,
|
| 1818 |
+
"step": 18100,
|
| 1819 |
+
"train_runtime": 10205.6407,
|
| 1820 |
+
"train_tokens_per_second": 116229.998
|
| 1821 |
+
},
|
| 1822 |
+
{
|
| 1823 |
+
"epoch": 0.182,
|
| 1824 |
+
"grad_norm": 0.2586776614189148,
|
| 1825 |
+
"learning_rate": 0.0002782083744808141,
|
| 1826 |
+
"loss": 0.6292,
|
| 1827 |
+
"num_input_tokens_seen": 1192755200,
|
| 1828 |
+
"step": 18200,
|
| 1829 |
+
"train_runtime": 10267.3133,
|
| 1830 |
+
"train_tokens_per_second": 116170.138
|
| 1831 |
+
},
|
| 1832 |
+
{
|
| 1833 |
+
"epoch": 0.183,
|
| 1834 |
+
"grad_norm": 0.6283107399940491,
|
| 1835 |
+
"learning_rate": 0.000277960645385489,
|
| 1836 |
+
"loss": 0.6364,
|
| 1837 |
+
"num_input_tokens_seen": 1199308800,
|
| 1838 |
+
"step": 18300,
|
| 1839 |
+
"train_runtime": 10323.5921,
|
| 1840 |
+
"train_tokens_per_second": 116171.657
|
| 1841 |
+
},
|
| 1842 |
+
{
|
| 1843 |
+
"epoch": 0.184,
|
| 1844 |
+
"grad_norm": 0.2882465124130249,
|
| 1845 |
+
"learning_rate": 0.00027771162772788544,
|
| 1846 |
+
"loss": 0.6318,
|
| 1847 |
+
"num_input_tokens_seen": 1205862400,
|
| 1848 |
+
"step": 18400,
|
| 1849 |
+
"train_runtime": 10377.5746,
|
| 1850 |
+
"train_tokens_per_second": 116198.866
|
| 1851 |
+
},
|
| 1852 |
+
{
|
| 1853 |
+
"epoch": 0.185,
|
| 1854 |
+
"grad_norm": 0.3091796934604645,
|
| 1855 |
+
"learning_rate": 0.00027746132401560857,
|
| 1856 |
+
"loss": 0.6269,
|
| 1857 |
+
"num_input_tokens_seen": 1212416000,
|
| 1858 |
+
"step": 18500,
|
| 1859 |
+
"train_runtime": 10432.3665,
|
| 1860 |
+
"train_tokens_per_second": 116216.776
|
| 1861 |
+
},
|
| 1862 |
+
{
|
| 1863 |
+
"epoch": 0.186,
|
| 1864 |
+
"grad_norm": 0.280862957239151,
|
| 1865 |
+
"learning_rate": 0.0002772097367692139,
|
| 1866 |
+
"loss": 0.6323,
|
| 1867 |
+
"num_input_tokens_seen": 1218969600,
|
| 1868 |
+
"step": 18600,
|
| 1869 |
+
"train_runtime": 10487.2106,
|
| 1870 |
+
"train_tokens_per_second": 116233.921
|
| 1871 |
+
},
|
| 1872 |
+
{
|
| 1873 |
+
"epoch": 0.187,
|
| 1874 |
+
"grad_norm": 0.24253763258457184,
|
| 1875 |
+
"learning_rate": 0.00027695686852218226,
|
| 1876 |
+
"loss": 0.6374,
|
| 1877 |
+
"num_input_tokens_seen": 1225523200,
|
| 1878 |
+
"step": 18700,
|
| 1879 |
+
"train_runtime": 10547.3913,
|
| 1880 |
+
"train_tokens_per_second": 116192.067
|
| 1881 |
+
},
|
| 1882 |
+
{
|
| 1883 |
+
"epoch": 0.188,
|
| 1884 |
+
"grad_norm": 0.26167231798171997,
|
| 1885 |
+
"learning_rate": 0.00027670272182089416,
|
| 1886 |
+
"loss": 0.6357,
|
| 1887 |
+
"num_input_tokens_seen": 1232076800,
|
| 1888 |
+
"step": 18800,
|
| 1889 |
+
"train_runtime": 10602.8292,
|
| 1890 |
+
"train_tokens_per_second": 116202.645
|
| 1891 |
+
},
|
| 1892 |
+
{
|
| 1893 |
+
"epoch": 0.189,
|
| 1894 |
+
"grad_norm": 0.31137940287590027,
|
| 1895 |
+
"learning_rate": 0.0002764472992246039,
|
| 1896 |
+
"loss": 0.6305,
|
| 1897 |
+
"num_input_tokens_seen": 1238630400,
|
| 1898 |
+
"step": 18900,
|
| 1899 |
+
"train_runtime": 10657.9196,
|
| 1900 |
+
"train_tokens_per_second": 116216.903
|
| 1901 |
+
},
|
| 1902 |
+
{
|
| 1903 |
+
"epoch": 0.19,
|
| 1904 |
+
"grad_norm": 0.2601984441280365,
|
| 1905 |
+
"learning_rate": 0.0002761906033054143,
|
| 1906 |
+
"loss": 0.6301,
|
| 1907 |
+
"num_input_tokens_seen": 1245184000,
|
| 1908 |
+
"step": 19000,
|
| 1909 |
+
"train_runtime": 10712.1812,
|
| 1910 |
+
"train_tokens_per_second": 116240.005
|
| 1911 |
+
},
|
| 1912 |
+
{
|
| 1913 |
+
"epoch": 0.191,
|
| 1914 |
+
"grad_norm": 0.42287951707839966,
|
| 1915 |
+
"learning_rate": 0.00027593263664825045,
|
| 1916 |
+
"loss": 0.6324,
|
| 1917 |
+
"num_input_tokens_seen": 1251737600,
|
| 1918 |
+
"step": 19100,
|
| 1919 |
+
"train_runtime": 10766.7423,
|
| 1920 |
+
"train_tokens_per_second": 116259.642
|
| 1921 |
+
},
|
| 1922 |
+
{
|
| 1923 |
+
"epoch": 0.192,
|
| 1924 |
+
"grad_norm": 0.28982725739479065,
|
| 1925 |
+
"learning_rate": 0.00027567340185083363,
|
| 1926 |
+
"loss": 0.6291,
|
| 1927 |
+
"num_input_tokens_seen": 1258291200,
|
| 1928 |
+
"step": 19200,
|
| 1929 |
+
"train_runtime": 10826.785,
|
| 1930 |
+
"train_tokens_per_second": 116220.208
|
| 1931 |
+
},
|
| 1932 |
+
{
|
| 1933 |
+
"epoch": 0.193,
|
| 1934 |
+
"grad_norm": 0.2896488308906555,
|
| 1935 |
+
"learning_rate": 0.00027541290152365537,
|
| 1936 |
+
"loss": 0.6354,
|
| 1937 |
+
"num_input_tokens_seen": 1264844800,
|
| 1938 |
+
"step": 19300,
|
| 1939 |
+
"train_runtime": 10881.8628,
|
| 1940 |
+
"train_tokens_per_second": 116234.216
|
| 1941 |
+
},
|
| 1942 |
+
{
|
| 1943 |
+
"epoch": 0.194,
|
| 1944 |
+
"grad_norm": 0.27890026569366455,
|
| 1945 |
+
"learning_rate": 0.00027515113828995117,
|
| 1946 |
+
"loss": 0.6339,
|
| 1947 |
+
"num_input_tokens_seen": 1271398400,
|
| 1948 |
+
"step": 19400,
|
| 1949 |
+
"train_runtime": 10937.449,
|
| 1950 |
+
"train_tokens_per_second": 116242.682
|
| 1951 |
+
},
|
| 1952 |
+
{
|
| 1953 |
+
"epoch": 0.195,
|
| 1954 |
+
"grad_norm": 0.2991676926612854,
|
| 1955 |
+
"learning_rate": 0.00027488811478567374,
|
| 1956 |
+
"loss": 0.6282,
|
| 1957 |
+
"num_input_tokens_seen": 1277952000,
|
| 1958 |
+
"step": 19500,
|
| 1959 |
+
"train_runtime": 10991.5778,
|
| 1960 |
+
"train_tokens_per_second": 116266.475
|
| 1961 |
+
},
|
| 1962 |
+
{
|
| 1963 |
+
"epoch": 0.196,
|
| 1964 |
+
"grad_norm": 0.2899467349052429,
|
| 1965 |
+
"learning_rate": 0.0002746238336594671,
|
| 1966 |
+
"loss": 0.6264,
|
| 1967 |
+
"num_input_tokens_seen": 1284505600,
|
| 1968 |
+
"step": 19600,
|
| 1969 |
+
"train_runtime": 11050.8685,
|
| 1970 |
+
"train_tokens_per_second": 116235.714
|
| 1971 |
+
},
|
| 1972 |
+
{
|
| 1973 |
+
"epoch": 0.197,
|
| 1974 |
+
"grad_norm": 0.32908034324645996,
|
| 1975 |
+
"learning_rate": 0.00027435829757263894,
|
| 1976 |
+
"loss": 0.6321,
|
| 1977 |
+
"num_input_tokens_seen": 1291059200,
|
| 1978 |
+
"step": 19700,
|
| 1979 |
+
"train_runtime": 11105.6721,
|
| 1980 |
+
"train_tokens_per_second": 116252.234
|
| 1981 |
+
},
|
| 1982 |
+
{
|
| 1983 |
+
"epoch": 0.198,
|
| 1984 |
+
"grad_norm": 0.23919178545475006,
|
| 1985 |
+
"learning_rate": 0.0002740915091991349,
|
| 1986 |
+
"loss": 0.6342,
|
| 1987 |
+
"num_input_tokens_seen": 1297612800,
|
| 1988 |
+
"step": 19800,
|
| 1989 |
+
"train_runtime": 11160.1183,
|
| 1990 |
+
"train_tokens_per_second": 116272.316
|
| 1991 |
+
},
|
| 1992 |
+
{
|
| 1993 |
+
"epoch": 0.199,
|
| 1994 |
+
"grad_norm": 0.5094599723815918,
|
| 1995 |
+
"learning_rate": 0.0002738234712255109,
|
| 1996 |
+
"loss": 0.6317,
|
| 1997 |
+
"num_input_tokens_seen": 1304166400,
|
| 1998 |
+
"step": 19900,
|
| 1999 |
+
"train_runtime": 11221.0166,
|
| 2000 |
+
"train_tokens_per_second": 116225.333
|
| 2001 |
+
},
|
| 2002 |
+
{
|
| 2003 |
+
"epoch": 0.2,
|
| 2004 |
+
"grad_norm": 0.37415555119514465,
|
| 2005 |
+
"learning_rate": 0.00027355418635090635,
|
| 2006 |
+
"loss": 0.634,
|
| 2007 |
+
"num_input_tokens_seen": 1310720000,
|
| 2008 |
+
"step": 20000,
|
| 2009 |
+
"train_runtime": 11276.9767,
|
| 2010 |
+
"train_tokens_per_second": 116229.734
|
| 2011 |
+
},
|
| 2012 |
+
{
|
| 2013 |
+
"epoch": 0.201,
|
| 2014 |
+
"grad_norm": 0.39444148540496826,
|
| 2015 |
+
"learning_rate": 0.000273283657287017,
|
| 2016 |
+
"loss": 0.6272,
|
| 2017 |
+
"num_input_tokens_seen": 1317273600,
|
| 2018 |
+
"step": 20100,
|
| 2019 |
+
"train_runtime": 11332.3114,
|
| 2020 |
+
"train_tokens_per_second": 116240.505
|
| 2021 |
+
},
|
| 2022 |
+
{
|
| 2023 |
+
"epoch": 0.202,
|
| 2024 |
+
"grad_norm": 0.2751108705997467,
|
| 2025 |
+
"learning_rate": 0.00027301188675806745,
|
| 2026 |
+
"loss": 0.6385,
|
| 2027 |
+
"num_input_tokens_seen": 1323827200,
|
| 2028 |
+
"step": 20200,
|
| 2029 |
+
"train_runtime": 11387.621,
|
| 2030 |
+
"train_tokens_per_second": 116251.427
|
| 2031 |
+
},
|
| 2032 |
+
{
|
| 2033 |
+
"epoch": 0.203,
|
| 2034 |
+
"grad_norm": 0.2579997479915619,
|
| 2035 |
+
"learning_rate": 0.0002727388775007839,
|
| 2036 |
+
"loss": 0.6278,
|
| 2037 |
+
"num_input_tokens_seen": 1330380800,
|
| 2038 |
+
"step": 20300,
|
| 2039 |
+
"train_runtime": 11441.506,
|
| 2040 |
+
"train_tokens_per_second": 116276.721
|
| 2041 |
+
},
|
| 2042 |
+
{
|
| 2043 |
+
"epoch": 0.204,
|
| 2044 |
+
"grad_norm": 0.27806708216667175,
|
| 2045 |
+
"learning_rate": 0.0002724646322643666,
|
| 2046 |
+
"loss": 0.6292,
|
| 2047 |
+
"num_input_tokens_seen": 1336934400,
|
| 2048 |
+
"step": 20400,
|
| 2049 |
+
"train_runtime": 11496.0966,
|
| 2050 |
+
"train_tokens_per_second": 116294.638
|
| 2051 |
+
},
|
| 2052 |
+
{
|
| 2053 |
+
"epoch": 0.205,
|
| 2054 |
+
"grad_norm": 0.48101556301116943,
|
| 2055 |
+
"learning_rate": 0.000272189153810462,
|
| 2056 |
+
"loss": 0.6335,
|
| 2057 |
+
"num_input_tokens_seen": 1343488000,
|
| 2058 |
+
"step": 20500,
|
| 2059 |
+
"train_runtime": 11557.1454,
|
| 2060 |
+
"train_tokens_per_second": 116247.391
|
| 2061 |
+
},
|
| 2062 |
+
{
|
| 2063 |
+
"epoch": 0.206,
|
| 2064 |
+
"grad_norm": 0.24878458678722382,
|
| 2065 |
+
"learning_rate": 0.0002719124449131351,
|
| 2066 |
+
"loss": 0.6309,
|
| 2067 |
+
"num_input_tokens_seen": 1350041600,
|
| 2068 |
+
"step": 20600,
|
| 2069 |
+
"train_runtime": 11613.0753,
|
| 2070 |
+
"train_tokens_per_second": 116251.86
|
| 2071 |
+
},
|
| 2072 |
+
{
|
| 2073 |
+
"epoch": 0.207,
|
| 2074 |
+
"grad_norm": 0.2999299466609955,
|
| 2075 |
+
"learning_rate": 0.00027163450835884144,
|
| 2076 |
+
"loss": 0.627,
|
| 2077 |
+
"num_input_tokens_seen": 1356595200,
|
| 2078 |
+
"step": 20700,
|
| 2079 |
+
"train_runtime": 11667.4657,
|
| 2080 |
+
"train_tokens_per_second": 116271.625
|
| 2081 |
+
},
|
| 2082 |
+
{
|
| 2083 |
+
"epoch": 0.208,
|
| 2084 |
+
"grad_norm": 0.2567562162876129,
|
| 2085 |
+
"learning_rate": 0.00027135534694639894,
|
| 2086 |
+
"loss": 0.6326,
|
| 2087 |
+
"num_input_tokens_seen": 1363148800,
|
| 2088 |
+
"step": 20800,
|
| 2089 |
+
"train_runtime": 11723.8934,
|
| 2090 |
+
"train_tokens_per_second": 116271.0
|
| 2091 |
+
},
|
| 2092 |
+
{
|
| 2093 |
+
"epoch": 0.209,
|
| 2094 |
+
"grad_norm": 0.3484431207180023,
|
| 2095 |
+
"learning_rate": 0.00027107496348696003,
|
| 2096 |
+
"loss": 0.6356,
|
| 2097 |
+
"num_input_tokens_seen": 1369702400,
|
| 2098 |
+
"step": 20900,
|
| 2099 |
+
"train_runtime": 11777.6307,
|
| 2100 |
+
"train_tokens_per_second": 116296.939
|
| 2101 |
+
},
|
| 2102 |
+
{
|
| 2103 |
+
"epoch": 0.21,
|
| 2104 |
+
"grad_norm": 0.23648300766944885,
|
| 2105 |
+
"learning_rate": 0.00027079336080398296,
|
| 2106 |
+
"loss": 0.6256,
|
| 2107 |
+
"num_input_tokens_seen": 1376256000,
|
| 2108 |
+
"step": 21000,
|
| 2109 |
+
"train_runtime": 11838.2167,
|
| 2110 |
+
"train_tokens_per_second": 116255.348
|
| 2111 |
+
},
|
| 2112 |
+
{
|
| 2113 |
+
"epoch": 0.211,
|
| 2114 |
+
"grad_norm": 0.33549532294273376,
|
| 2115 |
+
"learning_rate": 0.00027051054173320366,
|
| 2116 |
+
"loss": 0.6271,
|
| 2117 |
+
"num_input_tokens_seen": 1382809600,
|
| 2118 |
+
"step": 21100,
|
| 2119 |
+
"train_runtime": 11893.4157,
|
| 2120 |
+
"train_tokens_per_second": 116266.819
|
| 2121 |
+
},
|
| 2122 |
+
{
|
| 2123 |
+
"epoch": 0.212,
|
| 2124 |
+
"grad_norm": 0.2517142593860626,
|
| 2125 |
+
"learning_rate": 0.000270226509122607,
|
| 2126 |
+
"loss": 0.6254,
|
| 2127 |
+
"num_input_tokens_seen": 1389363200,
|
| 2128 |
+
"step": 21200,
|
| 2129 |
+
"train_runtime": 11947.4932,
|
| 2130 |
+
"train_tokens_per_second": 116289.098
|
| 2131 |
+
},
|
| 2132 |
+
{
|
| 2133 |
+
"epoch": 0.213,
|
| 2134 |
+
"grad_norm": 0.2526894807815552,
|
| 2135 |
+
"learning_rate": 0.0002699412658323983,
|
| 2136 |
+
"loss": 0.633,
|
| 2137 |
+
"num_input_tokens_seen": 1395916800,
|
| 2138 |
+
"step": 21300,
|
| 2139 |
+
"train_runtime": 12002.0117,
|
| 2140 |
+
"train_tokens_per_second": 116306.902
|
| 2141 |
+
},
|
| 2142 |
+
{
|
| 2143 |
+
"epoch": 0.214,
|
| 2144 |
+
"grad_norm": 0.283974289894104,
|
| 2145 |
+
"learning_rate": 0.00026965481473497423,
|
| 2146 |
+
"loss": 0.6273,
|
| 2147 |
+
"num_input_tokens_seen": 1402470400,
|
| 2148 |
+
"step": 21400,
|
| 2149 |
+
"train_runtime": 12063.5743,
|
| 2150 |
+
"train_tokens_per_second": 116256.622
|
| 2151 |
+
},
|
| 2152 |
+
{
|
| 2153 |
+
"epoch": 0.215,
|
| 2154 |
+
"grad_norm": 0.24663467705249786,
|
| 2155 |
+
"learning_rate": 0.0002693671587148942,
|
| 2156 |
+
"loss": 0.6236,
|
| 2157 |
+
"num_input_tokens_seen": 1409024000,
|
| 2158 |
+
"step": 21500,
|
| 2159 |
+
"train_runtime": 12118.0936,
|
| 2160 |
+
"train_tokens_per_second": 116274.395
|
| 2161 |
+
},
|
| 2162 |
+
{
|
| 2163 |
+
"epoch": 0.216,
|
| 2164 |
+
"grad_norm": 0.30240553617477417,
|
| 2165 |
+
"learning_rate": 0.0002690783006688511,
|
| 2166 |
+
"loss": 0.6272,
|
| 2167 |
+
"num_input_tokens_seen": 1415577600,
|
| 2168 |
+
"step": 21600,
|
| 2169 |
+
"train_runtime": 12174.4024,
|
| 2170 |
+
"train_tokens_per_second": 116274.915
|
| 2171 |
+
},
|
| 2172 |
+
{
|
| 2173 |
+
"epoch": 0.217,
|
| 2174 |
+
"grad_norm": 0.2617557644844055,
|
| 2175 |
+
"learning_rate": 0.0002687882435056423,
|
| 2176 |
+
"loss": 0.6256,
|
| 2177 |
+
"num_input_tokens_seen": 1422131200,
|
| 2178 |
+
"step": 21700,
|
| 2179 |
+
"train_runtime": 12229.4977,
|
| 2180 |
+
"train_tokens_per_second": 116286.967
|
| 2181 |
+
},
|
| 2182 |
+
{
|
| 2183 |
+
"epoch": 0.218,
|
| 2184 |
+
"grad_norm": 0.3469904661178589,
|
| 2185 |
+
"learning_rate": 0.0002684969901461402,
|
| 2186 |
+
"loss": 0.634,
|
| 2187 |
+
"num_input_tokens_seen": 1428684800,
|
| 2188 |
+
"step": 21800,
|
| 2189 |
+
"train_runtime": 12284.4989,
|
| 2190 |
+
"train_tokens_per_second": 116299.803
|
| 2191 |
+
},
|
| 2192 |
+
{
|
| 2193 |
+
"epoch": 0.219,
|
| 2194 |
+
"grad_norm": 0.32695531845092773,
|
| 2195 |
+
"learning_rate": 0.000268204543523263,
|
| 2196 |
+
"loss": 0.6343,
|
| 2197 |
+
"num_input_tokens_seen": 1435238400,
|
| 2198 |
+
"step": 21900,
|
| 2199 |
+
"train_runtime": 12340.2707,
|
| 2200 |
+
"train_tokens_per_second": 116305.261
|
| 2201 |
+
},
|
| 2202 |
+
{
|
| 2203 |
+
"epoch": 0.22,
|
| 2204 |
+
"grad_norm": 0.2909862995147705,
|
| 2205 |
+
"learning_rate": 0.0002679109065819447,
|
| 2206 |
+
"loss": 0.6275,
|
| 2207 |
+
"num_input_tokens_seen": 1441792000,
|
| 2208 |
+
"step": 22000,
|
| 2209 |
+
"train_runtime": 12395.2144,
|
| 2210 |
+
"train_tokens_per_second": 116318.44
|
| 2211 |
+
},
|
| 2212 |
+
{
|
| 2213 |
+
"epoch": 0.221,
|
| 2214 |
+
"grad_norm": 0.32462117075920105,
|
| 2215 |
+
"learning_rate": 0.0002676160822791062,
|
| 2216 |
+
"loss": 0.6265,
|
| 2217 |
+
"num_input_tokens_seen": 1448345600,
|
| 2218 |
+
"step": 22100,
|
| 2219 |
+
"train_runtime": 12456.5575,
|
| 2220 |
+
"train_tokens_per_second": 116271.738
|
| 2221 |
+
},
|
| 2222 |
+
{
|
| 2223 |
+
"epoch": 0.222,
|
| 2224 |
+
"grad_norm": 0.32560595870018005,
|
| 2225 |
+
"learning_rate": 0.00026732007358362496,
|
| 2226 |
+
"loss": 0.6434,
|
| 2227 |
+
"num_input_tokens_seen": 1454899200,
|
| 2228 |
+
"step": 22200,
|
| 2229 |
+
"train_runtime": 12512.3995,
|
| 2230 |
+
"train_tokens_per_second": 116276.594
|
| 2231 |
+
},
|
| 2232 |
+
{
|
| 2233 |
+
"epoch": 0.223,
|
| 2234 |
+
"grad_norm": 0.2810288369655609,
|
| 2235 |
+
"learning_rate": 0.0002670228834763052,
|
| 2236 |
+
"loss": 0.6289,
|
| 2237 |
+
"num_input_tokens_seen": 1461452800,
|
| 2238 |
+
"step": 22300,
|
| 2239 |
+
"train_runtime": 12567.5646,
|
| 2240 |
+
"train_tokens_per_second": 116287.67
|
| 2241 |
+
},
|
| 2242 |
+
{
|
| 2243 |
+
"epoch": 0.224,
|
| 2244 |
+
"grad_norm": 0.2596298158168793,
|
| 2245 |
+
"learning_rate": 0.00026672451494984804,
|
| 2246 |
+
"loss": 0.6287,
|
| 2247 |
+
"num_input_tokens_seen": 1468006400,
|
| 2248 |
+
"step": 22400,
|
| 2249 |
+
"train_runtime": 12622.4346,
|
| 2250 |
+
"train_tokens_per_second": 116301.367
|
| 2251 |
+
},
|
| 2252 |
+
{
|
| 2253 |
+
"epoch": 0.225,
|
| 2254 |
+
"grad_norm": 0.2850113809108734,
|
| 2255 |
+
"learning_rate": 0.0002664249710088213,
|
| 2256 |
+
"loss": 0.6242,
|
| 2257 |
+
"num_input_tokens_seen": 1474560000,
|
| 2258 |
+
"step": 22500,
|
| 2259 |
+
"train_runtime": 12676.5798,
|
| 2260 |
+
"train_tokens_per_second": 116321.596
|
| 2261 |
+
},
|
| 2262 |
+
{
|
| 2263 |
+
"epoch": 0.226,
|
| 2264 |
+
"grad_norm": 0.27683302760124207,
|
| 2265 |
+
"learning_rate": 0.00026612425466962893,
|
| 2266 |
+
"loss": 0.6206,
|
| 2267 |
+
"num_input_tokens_seen": 1481113600,
|
| 2268 |
+
"step": 22600,
|
| 2269 |
+
"train_runtime": 12737.6237,
|
| 2270 |
+
"train_tokens_per_second": 116278.643
|
| 2271 |
+
},
|
| 2272 |
+
{
|
| 2273 |
+
"epoch": 0.227,
|
| 2274 |
+
"grad_norm": 0.239657461643219,
|
| 2275 |
+
"learning_rate": 0.00026582236896048134,
|
| 2276 |
+
"loss": 0.6225,
|
| 2277 |
+
"num_input_tokens_seen": 1487667200,
|
| 2278 |
+
"step": 22700,
|
| 2279 |
+
"train_runtime": 12792.2969,
|
| 2280 |
+
"train_tokens_per_second": 116293.987
|
| 2281 |
+
},
|
| 2282 |
+
{
|
| 2283 |
+
"epoch": 0.228,
|
| 2284 |
+
"grad_norm": 0.23831160366535187,
|
| 2285 |
+
"learning_rate": 0.00026551931692136413,
|
| 2286 |
+
"loss": 0.6239,
|
| 2287 |
+
"num_input_tokens_seen": 1494220800,
|
| 2288 |
+
"step": 22800,
|
| 2289 |
+
"train_runtime": 12846.1745,
|
| 2290 |
+
"train_tokens_per_second": 116316.402
|
| 2291 |
+
},
|
| 2292 |
+
{
|
| 2293 |
+
"epoch": 0.229,
|
| 2294 |
+
"grad_norm": 0.34589797258377075,
|
| 2295 |
+
"learning_rate": 0.00026521510160400804,
|
| 2296 |
+
"loss": 0.6247,
|
| 2297 |
+
"num_input_tokens_seen": 1500774400,
|
| 2298 |
+
"step": 22900,
|
| 2299 |
+
"train_runtime": 12900.799,
|
| 2300 |
+
"train_tokens_per_second": 116331.895
|
| 2301 |
+
},
|
| 2302 |
+
{
|
| 2303 |
+
"epoch": 0.23,
|
| 2304 |
+
"grad_norm": 0.2497359812259674,
|
| 2305 |
+
"learning_rate": 0.00026490972607185793,
|
| 2306 |
+
"loss": 0.62,
|
| 2307 |
+
"num_input_tokens_seen": 1507328000,
|
| 2308 |
+
"step": 23000,
|
| 2309 |
+
"train_runtime": 12961.1601,
|
| 2310 |
+
"train_tokens_per_second": 116295.763
|
| 2311 |
+
},
|
| 2312 |
+
{
|
| 2313 |
+
"epoch": 0.231,
|
| 2314 |
+
"grad_norm": 0.24375373125076294,
|
| 2315 |
+
"learning_rate": 0.0002646031934000421,
|
| 2316 |
+
"loss": 0.6315,
|
| 2317 |
+
"num_input_tokens_seen": 1513881600,
|
| 2318 |
+
"step": 23100,
|
| 2319 |
+
"train_runtime": 13015.5663,
|
| 2320 |
+
"train_tokens_per_second": 116313.156
|
| 2321 |
+
},
|
| 2322 |
+
{
|
| 2323 |
+
"epoch": 0.232,
|
| 2324 |
+
"grad_norm": 0.26220017671585083,
|
| 2325 |
+
"learning_rate": 0.00026429550667534095,
|
| 2326 |
+
"loss": 0.6285,
|
| 2327 |
+
"num_input_tokens_seen": 1520435200,
|
| 2328 |
+
"step": 23200,
|
| 2329 |
+
"train_runtime": 13070.9925,
|
| 2330 |
+
"train_tokens_per_second": 116321.327
|
| 2331 |
+
},
|
| 2332 |
+
{
|
| 2333 |
+
"epoch": 0.233,
|
| 2334 |
+
"grad_norm": 0.3479808568954468,
|
| 2335 |
+
"learning_rate": 0.0002639866689961565,
|
| 2336 |
+
"loss": 0.6212,
|
| 2337 |
+
"num_input_tokens_seen": 1526988800,
|
| 2338 |
+
"step": 23300,
|
| 2339 |
+
"train_runtime": 13127.7973,
|
| 2340 |
+
"train_tokens_per_second": 116317.213
|
| 2341 |
+
},
|
| 2342 |
+
{
|
| 2343 |
+
"epoch": 0.234,
|
| 2344 |
+
"grad_norm": 0.2807121276855469,
|
| 2345 |
+
"learning_rate": 0.00026367668347248083,
|
| 2346 |
+
"loss": 0.6233,
|
| 2347 |
+
"num_input_tokens_seen": 1533542400,
|
| 2348 |
+
"step": 23400,
|
| 2349 |
+
"train_runtime": 13181.874,
|
| 2350 |
+
"train_tokens_per_second": 116337.206
|
| 2351 |
+
},
|
| 2352 |
+
{
|
| 2353 |
+
"epoch": 0.235,
|
| 2354 |
+
"grad_norm": 0.22995448112487793,
|
| 2355 |
+
"learning_rate": 0.0002633655532258646,
|
| 2356 |
+
"loss": 0.6322,
|
| 2357 |
+
"num_input_tokens_seen": 1540096000,
|
| 2358 |
+
"step": 23500,
|
| 2359 |
+
"train_runtime": 13243.5947,
|
| 2360 |
+
"train_tokens_per_second": 116289.877
|
| 2361 |
+
},
|
| 2362 |
+
{
|
| 2363 |
+
"epoch": 0.236,
|
| 2364 |
+
"grad_norm": 0.2977929413318634,
|
| 2365 |
+
"learning_rate": 0.000263053281389386,
|
| 2366 |
+
"loss": 0.6255,
|
| 2367 |
+
"num_input_tokens_seen": 1546649600,
|
| 2368 |
+
"step": 23600,
|
| 2369 |
+
"train_runtime": 13297.1642,
|
| 2370 |
+
"train_tokens_per_second": 116314.244
|
| 2371 |
+
},
|
| 2372 |
+
{
|
| 2373 |
+
"epoch": 0.237,
|
| 2374 |
+
"grad_norm": 0.2991272211074829,
|
| 2375 |
+
"learning_rate": 0.0002627398711076189,
|
| 2376 |
+
"loss": 0.6178,
|
| 2377 |
+
"num_input_tokens_seen": 1553203200,
|
| 2378 |
+
"step": 23700,
|
| 2379 |
+
"train_runtime": 13352.7857,
|
| 2380 |
+
"train_tokens_per_second": 116320.537
|
| 2381 |
+
},
|
| 2382 |
+
{
|
| 2383 |
+
"epoch": 0.238,
|
| 2384 |
+
"grad_norm": 0.26155802607536316,
|
| 2385 |
+
"learning_rate": 0.0002624253255366014,
|
| 2386 |
+
"loss": 0.6173,
|
| 2387 |
+
"num_input_tokens_seen": 1559756800,
|
| 2388 |
+
"step": 23800,
|
| 2389 |
+
"train_runtime": 13406.4815,
|
| 2390 |
+
"train_tokens_per_second": 116343.486
|
| 2391 |
+
},
|
| 2392 |
+
{
|
| 2393 |
+
"epoch": 0.239,
|
| 2394 |
+
"grad_norm": 0.3786696493625641,
|
| 2395 |
+
"learning_rate": 0.0002621096478438039,
|
| 2396 |
+
"loss": 0.6275,
|
| 2397 |
+
"num_input_tokens_seen": 1566310400,
|
| 2398 |
+
"step": 23900,
|
| 2399 |
+
"train_runtime": 13461.9758,
|
| 2400 |
+
"train_tokens_per_second": 116350.707
|
| 2401 |
+
},
|
| 2402 |
+
{
|
| 2403 |
+
"epoch": 0.24,
|
| 2404 |
+
"grad_norm": 0.347703754901886,
|
| 2405 |
+
"learning_rate": 0.00026179284120809727,
|
| 2406 |
+
"loss": 0.6317,
|
| 2407 |
+
"num_input_tokens_seen": 1572864000,
|
| 2408 |
+
"step": 24000,
|
| 2409 |
+
"train_runtime": 13518.3676,
|
| 2410 |
+
"train_tokens_per_second": 116350.143
|
| 2411 |
+
},
|
| 2412 |
+
{
|
| 2413 |
+
"epoch": 0.241,
|
| 2414 |
+
"grad_norm": 0.2794933021068573,
|
| 2415 |
+
"learning_rate": 0.0002614749088197208,
|
| 2416 |
+
"loss": 0.6213,
|
| 2417 |
+
"num_input_tokens_seen": 1579417600,
|
| 2418 |
+
"step": 24100,
|
| 2419 |
+
"train_runtime": 13573.1619,
|
| 2420 |
+
"train_tokens_per_second": 116363.277
|
| 2421 |
+
},
|
| 2422 |
+
{
|
| 2423 |
+
"epoch": 0.242,
|
| 2424 |
+
"grad_norm": 0.4467460811138153,
|
| 2425 |
+
"learning_rate": 0.00026115585388025015,
|
| 2426 |
+
"loss": 0.618,
|
| 2427 |
+
"num_input_tokens_seen": 1585971200,
|
| 2428 |
+
"step": 24200,
|
| 2429 |
+
"train_runtime": 13633.9103,
|
| 2430 |
+
"train_tokens_per_second": 116325.483
|
| 2431 |
+
},
|
| 2432 |
+
{
|
| 2433 |
+
"epoch": 0.243,
|
| 2434 |
+
"grad_norm": 0.3274633288383484,
|
| 2435 |
+
"learning_rate": 0.00026083567960256493,
|
| 2436 |
+
"loss": 0.6198,
|
| 2437 |
+
"num_input_tokens_seen": 1592524800,
|
| 2438 |
+
"step": 24300,
|
| 2439 |
+
"train_runtime": 13690.8223,
|
| 2440 |
+
"train_tokens_per_second": 116320.61
|
| 2441 |
+
},
|
| 2442 |
+
{
|
| 2443 |
+
"epoch": 0.244,
|
| 2444 |
+
"grad_norm": 0.26471319794654846,
|
| 2445 |
+
"learning_rate": 0.00026051438921081667,
|
| 2446 |
+
"loss": 0.621,
|
| 2447 |
+
"num_input_tokens_seen": 1599078400,
|
| 2448 |
+
"step": 24400,
|
| 2449 |
+
"train_runtime": 13746.056,
|
| 2450 |
+
"train_tokens_per_second": 116329.979
|
| 2451 |
+
},
|
| 2452 |
+
{
|
| 2453 |
+
"epoch": 0.245,
|
| 2454 |
+
"grad_norm": 0.29486411809921265,
|
| 2455 |
+
"learning_rate": 0.00026019198594039595,
|
| 2456 |
+
"loss": 0.6275,
|
| 2457 |
+
"num_input_tokens_seen": 1605632000,
|
| 2458 |
+
"step": 24500,
|
| 2459 |
+
"train_runtime": 13802.5927,
|
| 2460 |
+
"train_tokens_per_second": 116328.29
|
| 2461 |
+
},
|
| 2462 |
+
{
|
| 2463 |
+
"epoch": 0.246,
|
| 2464 |
+
"grad_norm": 0.31198298931121826,
|
| 2465 |
+
"learning_rate": 0.00025986847303790026,
|
| 2466 |
+
"loss": 0.6232,
|
| 2467 |
+
"num_input_tokens_seen": 1612185600,
|
| 2468 |
+
"step": 24600,
|
| 2469 |
+
"train_runtime": 13856.7259,
|
| 2470 |
+
"train_tokens_per_second": 116346.792
|
| 2471 |
+
},
|
| 2472 |
+
{
|
| 2473 |
+
"epoch": 0.247,
|
| 2474 |
+
"grad_norm": 0.28400614857673645,
|
| 2475 |
+
"learning_rate": 0.00025954385376110076,
|
| 2476 |
+
"loss": 0.6213,
|
| 2477 |
+
"num_input_tokens_seen": 1618739200,
|
| 2478 |
+
"step": 24700,
|
| 2479 |
+
"train_runtime": 13911.4019,
|
| 2480 |
+
"train_tokens_per_second": 116360.609
|
| 2481 |
+
},
|
| 2482 |
+
{
|
| 2483 |
+
"epoch": 0.248,
|
| 2484 |
+
"grad_norm": 0.24376364052295685,
|
| 2485 |
+
"learning_rate": 0.00025921813137891005,
|
| 2486 |
+
"loss": 0.6279,
|
| 2487 |
+
"num_input_tokens_seen": 1625292800,
|
| 2488 |
+
"step": 24800,
|
| 2489 |
+
"train_runtime": 13968.2353,
|
| 2490 |
+
"train_tokens_per_second": 116356.345
|
| 2491 |
+
},
|
| 2492 |
+
{
|
| 2493 |
+
"epoch": 0.249,
|
| 2494 |
+
"grad_norm": 0.28743863105773926,
|
| 2495 |
+
"learning_rate": 0.000258891309171349,
|
| 2496 |
+
"loss": 0.6238,
|
| 2497 |
+
"num_input_tokens_seen": 1631846400,
|
| 2498 |
+
"step": 24900,
|
| 2499 |
+
"train_runtime": 14022.4846,
|
| 2500 |
+
"train_tokens_per_second": 116373.557
|
| 2501 |
+
},
|
| 2502 |
+
{
|
| 2503 |
+
"epoch": 0.25,
|
| 2504 |
+
"grad_norm": 0.33175599575042725,
|
| 2505 |
+
"learning_rate": 0.00025856339042951344,
|
| 2506 |
+
"loss": 0.6159,
|
| 2507 |
+
"num_input_tokens_seen": 1638400000,
|
| 2508 |
+
"step": 25000,
|
| 2509 |
+
"train_runtime": 14076.9493,
|
| 2510 |
+
"train_tokens_per_second": 116388.855
|
| 2511 |
+
},
|
| 2512 |
+
{
|
| 2513 |
+
"epoch": 0.251,
|
| 2514 |
+
"grad_norm": 0.31394195556640625,
|
| 2515 |
+
"learning_rate": 0.0002582343784555415,
|
| 2516 |
+
"loss": 0.6194,
|
| 2517 |
+
"num_input_tokens_seen": 1644953600,
|
| 2518 |
+
"step": 25100,
|
| 2519 |
+
"train_runtime": 14137.6518,
|
| 2520 |
+
"train_tokens_per_second": 116352.675
|
| 2521 |
+
},
|
| 2522 |
+
{
|
| 2523 |
+
"epoch": 0.252,
|
| 2524 |
+
"grad_norm": 0.2717108428478241,
|
| 2525 |
+
"learning_rate": 0.00025790427656258017,
|
| 2526 |
+
"loss": 0.63,
|
| 2527 |
+
"num_input_tokens_seen": 1651507200,
|
| 2528 |
+
"step": 25200,
|
| 2529 |
+
"train_runtime": 14191.6974,
|
| 2530 |
+
"train_tokens_per_second": 116371.365
|
| 2531 |
+
},
|
| 2532 |
+
{
|
| 2533 |
+
"epoch": 0.253,
|
| 2534 |
+
"grad_norm": 0.2907046377658844,
|
| 2535 |
+
"learning_rate": 0.00025757308807475185,
|
| 2536 |
+
"loss": 0.6214,
|
| 2537 |
+
"num_input_tokens_seen": 1658060800,
|
| 2538 |
+
"step": 25300,
|
| 2539 |
+
"train_runtime": 14247.6012,
|
| 2540 |
+
"train_tokens_per_second": 116374.734
|
| 2541 |
+
},
|
| 2542 |
+
{
|
| 2543 |
+
"epoch": 0.254,
|
| 2544 |
+
"grad_norm": 0.25791919231414795,
|
| 2545 |
+
"learning_rate": 0.00025724081632712086,
|
| 2546 |
+
"loss": 0.6197,
|
| 2547 |
+
"num_input_tokens_seen": 1664614400,
|
| 2548 |
+
"step": 25400,
|
| 2549 |
+
"train_runtime": 14302.1699,
|
| 2550 |
+
"train_tokens_per_second": 116388.94
|
| 2551 |
+
},
|
| 2552 |
+
{
|
| 2553 |
+
"epoch": 0.255,
|
| 2554 |
+
"grad_norm": 0.23399506509304047,
|
| 2555 |
+
"learning_rate": 0.0002569074646656601,
|
| 2556 |
+
"loss": 0.6145,
|
| 2557 |
+
"num_input_tokens_seen": 1671168000,
|
| 2558 |
+
"step": 25500,
|
| 2559 |
+
"train_runtime": 14362.8828,
|
| 2560 |
+
"train_tokens_per_second": 116353.243
|
| 2561 |
+
},
|
| 2562 |
+
{
|
| 2563 |
+
"epoch": 0.256,
|
| 2564 |
+
"grad_norm": 0.3080284297466278,
|
| 2565 |
+
"learning_rate": 0.00025657303644721695,
|
| 2566 |
+
"loss": 0.6288,
|
| 2567 |
+
"num_input_tokens_seen": 1677721600,
|
| 2568 |
+
"step": 25600,
|
| 2569 |
+
"train_runtime": 14418.2746,
|
| 2570 |
+
"train_tokens_per_second": 116360.775
|
| 2571 |
+
},
|
| 2572 |
+
{
|
| 2573 |
+
"epoch": 0.257,
|
| 2574 |
+
"grad_norm": 0.27764490246772766,
|
| 2575 |
+
"learning_rate": 0.00025623753503948004,
|
| 2576 |
+
"loss": 0.6302,
|
| 2577 |
+
"num_input_tokens_seen": 1684275200,
|
| 2578 |
+
"step": 25700,
|
| 2579 |
+
"train_runtime": 14472.1462,
|
| 2580 |
+
"train_tokens_per_second": 116380.471
|
| 2581 |
+
},
|
| 2582 |
+
{
|
| 2583 |
+
"epoch": 0.258,
|
| 2584 |
+
"grad_norm": 0.3530017137527466,
|
| 2585 |
+
"learning_rate": 0.00025590096382094475,
|
| 2586 |
+
"loss": 0.6205,
|
| 2587 |
+
"num_input_tokens_seen": 1690828800,
|
| 2588 |
+
"step": 25800,
|
| 2589 |
+
"train_runtime": 14527.093,
|
| 2590 |
+
"train_tokens_per_second": 116391.407
|
| 2591 |
+
},
|
| 2592 |
+
{
|
| 2593 |
+
"epoch": 0.259,
|
| 2594 |
+
"grad_norm": 0.28179532289505005,
|
| 2595 |
+
"learning_rate": 0.00025556332618087945,
|
| 2596 |
+
"loss": 0.6196,
|
| 2597 |
+
"num_input_tokens_seen": 1697382400,
|
| 2598 |
+
"step": 25900,
|
| 2599 |
+
"train_runtime": 14586.2883,
|
| 2600 |
+
"train_tokens_per_second": 116368.357
|
| 2601 |
+
},
|
| 2602 |
+
{
|
| 2603 |
+
"epoch": 0.26,
|
| 2604 |
+
"grad_norm": 0.27701336145401,
|
| 2605 |
+
"learning_rate": 0.00025522462551929155,
|
| 2606 |
+
"loss": 0.6244,
|
| 2607 |
+
"num_input_tokens_seen": 1703936000,
|
| 2608 |
+
"step": 26000,
|
| 2609 |
+
"train_runtime": 14641.6003,
|
| 2610 |
+
"train_tokens_per_second": 116376.35
|
| 2611 |
+
},
|
| 2612 |
+
{
|
| 2613 |
+
"epoch": 0.261,
|
| 2614 |
+
"grad_norm": 0.2609594464302063,
|
| 2615 |
+
"learning_rate": 0.00025488486524689283,
|
| 2616 |
+
"loss": 0.625,
|
| 2617 |
+
"num_input_tokens_seen": 1710489600,
|
| 2618 |
+
"step": 26100,
|
| 2619 |
+
"train_runtime": 14698.0992,
|
| 2620 |
+
"train_tokens_per_second": 116374.884
|
| 2621 |
+
},
|
| 2622 |
+
{
|
| 2623 |
+
"epoch": 0.262,
|
| 2624 |
+
"grad_norm": 0.2739205062389374,
|
| 2625 |
+
"learning_rate": 0.00025454404878506555,
|
| 2626 |
+
"loss": 0.6214,
|
| 2627 |
+
"num_input_tokens_seen": 1717043200,
|
| 2628 |
+
"step": 26200,
|
| 2629 |
+
"train_runtime": 14752.0923,
|
| 2630 |
+
"train_tokens_per_second": 116393.198
|
| 2631 |
+
},
|
| 2632 |
+
{
|
| 2633 |
+
"epoch": 0.263,
|
| 2634 |
+
"grad_norm": 0.27936309576034546,
|
| 2635 |
+
"learning_rate": 0.0002542021795658276,
|
| 2636 |
+
"loss": 0.6203,
|
| 2637 |
+
"num_input_tokens_seen": 1723596800,
|
| 2638 |
+
"step": 26300,
|
| 2639 |
+
"train_runtime": 14806.7928,
|
| 2640 |
+
"train_tokens_per_second": 116405.816
|
| 2641 |
+
},
|
| 2642 |
+
{
|
| 2643 |
+
"epoch": 0.264,
|
| 2644 |
+
"grad_norm": 0.24206426739692688,
|
| 2645 |
+
"learning_rate": 0.0002538592610317984,
|
| 2646 |
+
"loss": 0.622,
|
| 2647 |
+
"num_input_tokens_seen": 1730150400,
|
| 2648 |
+
"step": 26400,
|
| 2649 |
+
"train_runtime": 14868.0925,
|
| 2650 |
+
"train_tokens_per_second": 116366.669
|
| 2651 |
+
},
|
| 2652 |
+
{
|
| 2653 |
+
"epoch": 0.265,
|
| 2654 |
+
"grad_norm": 0.29860714077949524,
|
| 2655 |
+
"learning_rate": 0.00025351529663616355,
|
| 2656 |
+
"loss": 0.6244,
|
| 2657 |
+
"num_input_tokens_seen": 1736704000,
|
| 2658 |
+
"step": 26500,
|
| 2659 |
+
"train_runtime": 14923.7941,
|
| 2660 |
+
"train_tokens_per_second": 116371.479
|
| 2661 |
+
},
|
| 2662 |
+
{
|
| 2663 |
+
"epoch": 0.266,
|
| 2664 |
+
"grad_norm": 0.2928908169269562,
|
| 2665 |
+
"learning_rate": 0.00025317028984264087,
|
| 2666 |
+
"loss": 0.6183,
|
| 2667 |
+
"num_input_tokens_seen": 1743257600,
|
| 2668 |
+
"step": 26600,
|
| 2669 |
+
"train_runtime": 14978.8307,
|
| 2670 |
+
"train_tokens_per_second": 116381.421
|
| 2671 |
+
},
|
| 2672 |
+
{
|
| 2673 |
+
"epoch": 0.267,
|
| 2674 |
+
"grad_norm": 0.32871657609939575,
|
| 2675 |
+
"learning_rate": 0.0002528242441254448,
|
| 2676 |
+
"loss": 0.6163,
|
| 2677 |
+
"num_input_tokens_seen": 1749811200,
|
| 2678 |
+
"step": 26700,
|
| 2679 |
+
"train_runtime": 15032.6295,
|
| 2680 |
+
"train_tokens_per_second": 116400.873
|
| 2681 |
+
},
|
| 2682 |
+
{
|
| 2683 |
+
"epoch": 0.268,
|
| 2684 |
+
"grad_norm": 0.26268327236175537,
|
| 2685 |
+
"learning_rate": 0.000252477162969252,
|
| 2686 |
+
"loss": 0.6211,
|
| 2687 |
+
"num_input_tokens_seen": 1756364800,
|
| 2688 |
+
"step": 26800,
|
| 2689 |
+
"train_runtime": 15087.2644,
|
| 2690 |
+
"train_tokens_per_second": 116413.735
|
| 2691 |
+
},
|
| 2692 |
+
{
|
| 2693 |
+
"epoch": 0.269,
|
| 2694 |
+
"grad_norm": 0.248652845621109,
|
| 2695 |
+
"learning_rate": 0.00025212904986916584,
|
| 2696 |
+
"loss": 0.6231,
|
| 2697 |
+
"num_input_tokens_seen": 1762918400,
|
| 2698 |
+
"step": 26900,
|
| 2699 |
+
"train_runtime": 15142.5638,
|
| 2700 |
+
"train_tokens_per_second": 116421.395
|
| 2701 |
+
},
|
| 2702 |
+
{
|
| 2703 |
+
"epoch": 0.27,
|
| 2704 |
+
"grad_norm": 0.28964775800704956,
|
| 2705 |
+
"learning_rate": 0.00025177990833068133,
|
| 2706 |
+
"loss": 0.6233,
|
| 2707 |
+
"num_input_tokens_seen": 1769472000,
|
| 2708 |
+
"step": 27000,
|
| 2709 |
+
"train_runtime": 15204.3907,
|
| 2710 |
+
"train_tokens_per_second": 116379.014
|
| 2711 |
+
},
|
| 2712 |
+
{
|
| 2713 |
+
"epoch": 0.271,
|
| 2714 |
+
"grad_norm": 0.3479403555393219,
|
| 2715 |
+
"learning_rate": 0.0002514297418696499,
|
| 2716 |
+
"loss": 0.6136,
|
| 2717 |
+
"num_input_tokens_seen": 1776025600,
|
| 2718 |
+
"step": 27100,
|
| 2719 |
+
"train_runtime": 15259.9986,
|
| 2720 |
+
"train_tokens_per_second": 116384.388
|
| 2721 |
+
},
|
| 2722 |
+
{
|
| 2723 |
+
"epoch": 0.272,
|
| 2724 |
+
"grad_norm": 0.2727760076522827,
|
| 2725 |
+
"learning_rate": 0.0002510785540122439,
|
| 2726 |
+
"loss": 0.6215,
|
| 2727 |
+
"num_input_tokens_seen": 1782579200,
|
| 2728 |
+
"step": 27200,
|
| 2729 |
+
"train_runtime": 15313.7249,
|
| 2730 |
+
"train_tokens_per_second": 116404.024
|
| 2731 |
+
},
|
| 2732 |
+
{
|
| 2733 |
+
"epoch": 0.273,
|
| 2734 |
+
"grad_norm": 0.2885916531085968,
|
| 2735 |
+
"learning_rate": 0.0002507263482949212,
|
| 2736 |
+
"loss": 0.6269,
|
| 2737 |
+
"num_input_tokens_seen": 1789132800,
|
| 2738 |
+
"step": 27300,
|
| 2739 |
+
"train_runtime": 15370.4778,
|
| 2740 |
+
"train_tokens_per_second": 116400.598
|
| 2741 |
+
},
|
| 2742 |
+
{
|
| 2743 |
+
"epoch": 0.274,
|
| 2744 |
+
"grad_norm": 0.2588540017604828,
|
| 2745 |
+
"learning_rate": 0.0002503731282643894,
|
| 2746 |
+
"loss": 0.619,
|
| 2747 |
+
"num_input_tokens_seen": 1795686400,
|
| 2748 |
+
"step": 27400,
|
| 2749 |
+
"train_runtime": 15425.7714,
|
| 2750 |
+
"train_tokens_per_second": 116408.208
|
| 2751 |
+
},
|
| 2752 |
+
{
|
| 2753 |
+
"epoch": 0.275,
|
| 2754 |
+
"grad_norm": 0.29681408405303955,
|
| 2755 |
+
"learning_rate": 0.0002500188974775704,
|
| 2756 |
+
"loss": 0.6175,
|
| 2757 |
+
"num_input_tokens_seen": 1802240000,
|
| 2758 |
+
"step": 27500,
|
| 2759 |
+
"train_runtime": 15480.68,
|
| 2760 |
+
"train_tokens_per_second": 116418.658
|
| 2761 |
+
},
|
| 2762 |
+
{
|
| 2763 |
+
"epoch": 0.276,
|
| 2764 |
+
"grad_norm": 0.319444864988327,
|
| 2765 |
+
"learning_rate": 0.00024966365950156416,
|
| 2766 |
+
"loss": 0.6155,
|
| 2767 |
+
"num_input_tokens_seen": 1808793600,
|
| 2768 |
+
"step": 27600,
|
| 2769 |
+
"train_runtime": 15541.0108,
|
| 2770 |
+
"train_tokens_per_second": 116388.414
|
| 2771 |
+
},
|
| 2772 |
+
{
|
| 2773 |
+
"epoch": 0.277,
|
| 2774 |
+
"grad_norm": 0.3779368996620178,
|
| 2775 |
+
"learning_rate": 0.00024930741791361326,
|
| 2776 |
+
"loss": 0.6238,
|
| 2777 |
+
"num_input_tokens_seen": 1815347200,
|
| 2778 |
+
"step": 27700,
|
| 2779 |
+
"train_runtime": 15597.4681,
|
| 2780 |
+
"train_tokens_per_second": 116387.3
|
| 2781 |
+
},
|
| 2782 |
+
{
|
| 2783 |
+
"epoch": 0.278,
|
| 2784 |
+
"grad_norm": 0.4451746940612793,
|
| 2785 |
+
"learning_rate": 0.0002489501763010664,
|
| 2786 |
+
"loss": 0.6201,
|
| 2787 |
+
"num_input_tokens_seen": 1821900800,
|
| 2788 |
+
"step": 27800,
|
| 2789 |
+
"train_runtime": 15648.3014,
|
| 2790 |
+
"train_tokens_per_second": 116428.024
|
| 2791 |
+
},
|
| 2792 |
+
{
|
| 2793 |
+
"epoch": 0.279,
|
| 2794 |
+
"grad_norm": 0.35030296444892883,
|
| 2795 |
+
"learning_rate": 0.00024859193826134285,
|
| 2796 |
+
"loss": 0.6173,
|
| 2797 |
+
"num_input_tokens_seen": 1828454400,
|
| 2798 |
+
"step": 27900,
|
| 2799 |
+
"train_runtime": 15708.8593,
|
| 2800 |
+
"train_tokens_per_second": 116396.383
|
| 2801 |
+
},
|
| 2802 |
+
{
|
| 2803 |
+
"epoch": 0.28,
|
| 2804 |
+
"grad_norm": 0.2962912321090698,
|
| 2805 |
+
"learning_rate": 0.00024823270740189556,
|
| 2806 |
+
"loss": 0.6154,
|
| 2807 |
+
"num_input_tokens_seen": 1835008000,
|
| 2808 |
+
"step": 28000,
|
| 2809 |
+
"train_runtime": 15764.0565,
|
| 2810 |
+
"train_tokens_per_second": 116404.556
|
| 2811 |
+
},
|
| 2812 |
+
{
|
| 2813 |
+
"epoch": 0.281,
|
| 2814 |
+
"grad_norm": 0.2939753234386444,
|
| 2815 |
+
"learning_rate": 0.00024787248734017527,
|
| 2816 |
+
"loss": 0.6222,
|
| 2817 |
+
"num_input_tokens_seen": 1841561600,
|
| 2818 |
+
"step": 28100,
|
| 2819 |
+
"train_runtime": 15818.596,
|
| 2820 |
+
"train_tokens_per_second": 116417.513
|
| 2821 |
+
},
|
| 2822 |
+
{
|
| 2823 |
+
"epoch": 0.282,
|
| 2824 |
+
"grad_norm": 0.24470455944538116,
|
| 2825 |
+
"learning_rate": 0.0002475112817035941,
|
| 2826 |
+
"loss": 0.6238,
|
| 2827 |
+
"num_input_tokens_seen": 1848115200,
|
| 2828 |
+
"step": 28200,
|
| 2829 |
+
"train_runtime": 15872.9134,
|
| 2830 |
+
"train_tokens_per_second": 116432.01
|
| 2831 |
+
},
|
| 2832 |
+
{
|
| 2833 |
+
"epoch": 0.283,
|
| 2834 |
+
"grad_norm": 0.31226930022239685,
|
| 2835 |
+
"learning_rate": 0.0002471490941294887,
|
| 2836 |
+
"loss": 0.6212,
|
| 2837 |
+
"num_input_tokens_seen": 1854668800,
|
| 2838 |
+
"step": 28300,
|
| 2839 |
+
"train_runtime": 15933.0328,
|
| 2840 |
+
"train_tokens_per_second": 116404.003
|
| 2841 |
+
},
|
| 2842 |
+
{
|
| 2843 |
+
"epoch": 0.284,
|
| 2844 |
+
"grad_norm": 0.2585756182670593,
|
| 2845 |
+
"learning_rate": 0.000246785928265084,
|
| 2846 |
+
"loss": 0.6191,
|
| 2847 |
+
"num_input_tokens_seen": 1861222400,
|
| 2848 |
+
"step": 28400,
|
| 2849 |
+
"train_runtime": 15987.3279,
|
| 2850 |
+
"train_tokens_per_second": 116418.604
|
| 2851 |
+
},
|
| 2852 |
+
{
|
| 2853 |
+
"epoch": 0.285,
|
| 2854 |
+
"grad_norm": 0.4430062770843506,
|
| 2855 |
+
"learning_rate": 0.0002464217877674562,
|
| 2856 |
+
"loss": 0.6249,
|
| 2857 |
+
"num_input_tokens_seen": 1867776000,
|
| 2858 |
+
"step": 28500,
|
| 2859 |
+
"train_runtime": 16041.837,
|
| 2860 |
+
"train_tokens_per_second": 116431.553
|
| 2861 |
+
},
|
| 2862 |
+
{
|
| 2863 |
+
"epoch": 0.286,
|
| 2864 |
+
"grad_norm": 0.31980106234550476,
|
| 2865 |
+
"learning_rate": 0.0002460566763034961,
|
| 2866 |
+
"loss": 0.6303,
|
| 2867 |
+
"num_input_tokens_seen": 1874329600,
|
| 2868 |
+
"step": 28600,
|
| 2869 |
+
"train_runtime": 16098.6135,
|
| 2870 |
+
"train_tokens_per_second": 116428.014
|
| 2871 |
+
},
|
| 2872 |
+
{
|
| 2873 |
+
"epoch": 0.287,
|
| 2874 |
+
"grad_norm": 0.26904401183128357,
|
| 2875 |
+
"learning_rate": 0.00024569059754987196,
|
| 2876 |
+
"loss": 0.6214,
|
| 2877 |
+
"num_input_tokens_seen": 1880883200,
|
| 2878 |
+
"step": 28700,
|
| 2879 |
+
"train_runtime": 16159.9296,
|
| 2880 |
+
"train_tokens_per_second": 116391.794
|
| 2881 |
+
},
|
| 2882 |
+
{
|
| 2883 |
+
"epoch": 0.288,
|
| 2884 |
+
"grad_norm": 0.2635185420513153,
|
| 2885 |
+
"learning_rate": 0.00024532355519299296,
|
| 2886 |
+
"loss": 0.6181,
|
| 2887 |
+
"num_input_tokens_seen": 1887436800,
|
| 2888 |
+
"step": 28800,
|
| 2889 |
+
"train_runtime": 16214.468,
|
| 2890 |
+
"train_tokens_per_second": 116404.485
|
| 2891 |
+
},
|
| 2892 |
+
{
|
| 2893 |
+
"epoch": 0.289,
|
| 2894 |
+
"grad_norm": 1.0330250263214111,
|
| 2895 |
+
"learning_rate": 0.0002449555529289714,
|
| 2896 |
+
"loss": 0.6241,
|
| 2897 |
+
"num_input_tokens_seen": 1893990400,
|
| 2898 |
+
"step": 28900,
|
| 2899 |
+
"train_runtime": 16268.5971,
|
| 2900 |
+
"train_tokens_per_second": 116420.02
|
| 2901 |
+
},
|
| 2902 |
+
{
|
| 2903 |
+
"epoch": 0.29,
|
| 2904 |
+
"grad_norm": 0.3014289438724518,
|
| 2905 |
+
"learning_rate": 0.0002445865944635861,
|
| 2906 |
+
"loss": 0.6295,
|
| 2907 |
+
"num_input_tokens_seen": 1900544000,
|
| 2908 |
+
"step": 29000,
|
| 2909 |
+
"train_runtime": 16324.9935,
|
| 2910 |
+
"train_tokens_per_second": 116419.281
|
| 2911 |
+
},
|
| 2912 |
+
{
|
| 2913 |
+
"epoch": 0.291,
|
| 2914 |
+
"grad_norm": 0.2632603347301483,
|
| 2915 |
+
"learning_rate": 0.0002442166835122446,
|
| 2916 |
+
"loss": 0.6188,
|
| 2917 |
+
"num_input_tokens_seen": 1907097600,
|
| 2918 |
+
"step": 29100,
|
| 2919 |
+
"train_runtime": 16378.6859,
|
| 2920 |
+
"train_tokens_per_second": 116437.766
|
| 2921 |
+
},
|
| 2922 |
+
{
|
| 2923 |
+
"epoch": 0.292,
|
| 2924 |
+
"grad_norm": 0.22746869921684265,
|
| 2925 |
+
"learning_rate": 0.00024384582379994614,
|
| 2926 |
+
"loss": 0.6172,
|
| 2927 |
+
"num_input_tokens_seen": 1913651200,
|
| 2928 |
+
"step": 29200,
|
| 2929 |
+
"train_runtime": 16434.3593,
|
| 2930 |
+
"train_tokens_per_second": 116442.093
|
| 2931 |
+
},
|
| 2932 |
+
{
|
| 2933 |
+
"epoch": 0.293,
|
| 2934 |
+
"grad_norm": 0.26981279253959656,
|
| 2935 |
+
"learning_rate": 0.00024347401906124388,
|
| 2936 |
+
"loss": 0.6161,
|
| 2937 |
+
"num_input_tokens_seen": 1920204800,
|
| 2938 |
+
"step": 29300,
|
| 2939 |
+
"train_runtime": 16494.2752,
|
| 2940 |
+
"train_tokens_per_second": 116416.44
|
| 2941 |
+
},
|
| 2942 |
+
{
|
| 2943 |
+
"epoch": 0.294,
|
| 2944 |
+
"grad_norm": 0.30709710717201233,
|
| 2945 |
+
"learning_rate": 0.0002431012730402075,
|
| 2946 |
+
"loss": 0.6225,
|
| 2947 |
+
"num_input_tokens_seen": 1926758400,
|
| 2948 |
+
"step": 29400,
|
| 2949 |
+
"train_runtime": 16549.5618,
|
| 2950 |
+
"train_tokens_per_second": 116423.53
|
| 2951 |
+
},
|
| 2952 |
+
{
|
| 2953 |
+
"epoch": 0.295,
|
| 2954 |
+
"grad_norm": 0.27512043714523315,
|
| 2955 |
+
"learning_rate": 0.00024272758949038517,
|
| 2956 |
+
"loss": 0.6167,
|
| 2957 |
+
"num_input_tokens_seen": 1933312000,
|
| 2958 |
+
"step": 29500,
|
| 2959 |
+
"train_runtime": 16604.0285,
|
| 2960 |
+
"train_tokens_per_second": 116436.321
|
| 2961 |
+
},
|
| 2962 |
+
{
|
| 2963 |
+
"epoch": 0.296,
|
| 2964 |
+
"grad_norm": 0.33047741651535034,
|
| 2965 |
+
"learning_rate": 0.00024235297217476616,
|
| 2966 |
+
"loss": 0.6194,
|
| 2967 |
+
"num_input_tokens_seen": 1939865600,
|
| 2968 |
+
"step": 29600,
|
| 2969 |
+
"train_runtime": 16665.4033,
|
| 2970 |
+
"train_tokens_per_second": 116400.76
|
| 2971 |
+
},
|
| 2972 |
+
{
|
| 2973 |
+
"epoch": 0.297,
|
| 2974 |
+
"grad_norm": 0.23181872069835663,
|
| 2975 |
+
"learning_rate": 0.00024197742486574268,
|
| 2976 |
+
"loss": 0.6235,
|
| 2977 |
+
"num_input_tokens_seen": 1946419200,
|
| 2978 |
+
"step": 29700,
|
| 2979 |
+
"train_runtime": 16719.5543,
|
| 2980 |
+
"train_tokens_per_second": 116415.735
|
| 2981 |
+
},
|
| 2982 |
+
{
|
| 2983 |
+
"epoch": 0.298,
|
| 2984 |
+
"grad_norm": 0.25960394740104675,
|
| 2985 |
+
"learning_rate": 0.0002416009513450719,
|
| 2986 |
+
"loss": 0.6189,
|
| 2987 |
+
"num_input_tokens_seen": 1952972800,
|
| 2988 |
+
"step": 29800,
|
| 2989 |
+
"train_runtime": 16775.5144,
|
| 2990 |
+
"train_tokens_per_second": 116418.058
|
| 2991 |
+
},
|
| 2992 |
+
{
|
| 2993 |
+
"epoch": 0.299,
|
| 2994 |
+
"grad_norm": 0.3980496823787689,
|
| 2995 |
+
"learning_rate": 0.00024122355540383806,
|
| 2996 |
+
"loss": 0.6205,
|
| 2997 |
+
"num_input_tokens_seen": 1959526400,
|
| 2998 |
+
"step": 29900,
|
| 2999 |
+
"train_runtime": 16831.9075,
|
| 3000 |
+
"train_tokens_per_second": 116417.369
|
| 3001 |
+
},
|
| 3002 |
+
{
|
| 3003 |
+
"epoch": 0.3,
|
| 3004 |
+
"grad_norm": 0.274140328168869,
|
| 3005 |
+
"learning_rate": 0.00024084524084241405,
|
| 3006 |
+
"loss": 0.6137,
|
| 3007 |
+
"num_input_tokens_seen": 1966080000,
|
| 3008 |
+
"step": 30000,
|
| 3009 |
+
"train_runtime": 16885.9089,
|
| 3010 |
+
"train_tokens_per_second": 116433.176
|
| 3011 |
+
}
|
| 3012 |
+
],
|
| 3013 |
+
"logging_steps": 100,
|
| 3014 |
+
"max_steps": 100000,
|
| 3015 |
+
"num_input_tokens_seen": 1966080000,
|
| 3016 |
+
"num_train_epochs": 9223372036854775807,
|
| 3017 |
+
"save_steps": 1000,
|
| 3018 |
+
"stateful_callbacks": {
|
| 3019 |
+
"TrainerControl": {
|
| 3020 |
+
"args": {
|
| 3021 |
+
"should_epoch_stop": false,
|
| 3022 |
+
"should_evaluate": false,
|
| 3023 |
+
"should_log": false,
|
| 3024 |
+
"should_save": true,
|
| 3025 |
+
"should_training_stop": false
|
| 3026 |
+
},
|
| 3027 |
+
"attributes": {}
|
| 3028 |
+
}
|
| 3029 |
+
},
|
| 3030 |
+
"total_flos": 3.755848237056e+16,
|
| 3031 |
+
"train_batch_size": 256,
|
| 3032 |
+
"trial_name": null,
|
| 3033 |
+
"trial_params": null
|
| 3034 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:28f66574fb7b259c84e9bba3af9a920cbbf76532514cd1149e468db46344d8d2
|
| 3 |
+
size 5841
|