Hugging Face's logo

fromziro
/
DataBench

Safetensors
English
ablations
datasets
1m
llama
custom_tokenizer
Model card Files
xet
 Community
Harley-ml commited on
Commit
3fb2833
·
verified ·
1 Parent(s): 173b8bc

Upload 3 files

Browse files
Files changed (3) hide show
  1. eot-2004-raw/config.json +32 -0
  2. eot-2004-raw/model.safetensors +3 -0
  3. eot-2004-raw/trainer_state.json +242 -0
eot-2004-raw/config.json ADDED
@@ -0,0 +1,32 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "LlamaForCausalLM"
4
+ ],
5
+ "attention_bias": false,
6
+ "attention_dropout": 0.0,
7
+ "bos_token_id": 1,
8
+ "dtype": "float32",
9
+ "eos_token_id": 2,
10
+ "head_dim": 32,
11
+ "hidden_act": "silu",
12
+ "hidden_size": 128,
13
+ "initializer_range": 0.02,
14
+ "intermediate_size": 344,
15
+ "max_position_embeddings": 256,
16
+ "mlp_bias": false,
17
+ "model_type": "llama",
18
+ "num_attention_heads": 4,
19
+ "num_hidden_layers": 8,
20
+ "num_key_value_heads": 2,
21
+ "pad_token_id": null,
22
+ "pretraining_tp": 1,
23
+ "rms_norm_eps": 1e-06,
24
+ "rope_parameters": {
25
+ "rope_theta": 2500.0,
26
+ "rope_type": "default"
27
+ },
28
+ "tie_word_embeddings": true,
29
+ "transformers_version": "5.13.1",
30
+ "use_cache": false,
31
+ "vocab_size": 2564
32
+ }
eot-2004-raw/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cfe25e48b6e2d8d89a4d0835535b42a8b37e7bc7c050a41bdbb72adeda6fa4ac
3
+ size 7129288
eot-2004-raw/trainer_state.json ADDED
@@ -0,0 +1,242 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 1.0,
6
+ "eval_steps": 500,
7
+ "global_step": 2428,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.0411946446961895,
14
+ "grad_norm": 0.5167902112007141,
15
+ "learning_rate": 0.003,
16
+ "loss": 5.962776489257813,
17
+ "step": 100
18
+ },
19
+ {
20
+ "epoch": 0.082389289392379,
21
+ "grad_norm": 0.6578283905982971,
22
+ "learning_rate": 0.003,
23
+ "loss": 4.3772125244140625,
24
+ "step": 200
25
+ },
26
+ {
27
+ "epoch": 0.12358393408856849,
28
+ "grad_norm": 0.7207351326942444,
29
+ "learning_rate": 0.003,
30
+ "loss": 3.604791564941406,
31
+ "step": 300
32
+ },
33
+ {
34
+ "epoch": 0.164778578784758,
35
+ "grad_norm": 0.49143338203430176,
36
+ "learning_rate": 0.003,
37
+ "loss": 3.2285305786132814,
38
+ "step": 400
39
+ },
40
+ {
41
+ "epoch": 0.2059732234809475,
42
+ "grad_norm": 0.5450193881988525,
43
+ "learning_rate": 0.003,
44
+ "loss": 2.997611083984375,
45
+ "step": 500
46
+ },
47
+ {
48
+ "epoch": 0.2059732234809475,
49
+ "eval_loss": 2.9541826248168945,
50
+ "eval_runtime": 8.7168,
51
+ "eval_samples_per_second": 368.256,
52
+ "eval_steps_per_second": 46.118,
53
+ "step": 500
54
+ },
55
+ {
56
+ "epoch": 0.24716786817713698,
57
+ "grad_norm": 0.3710590600967407,
58
+ "learning_rate": 0.003,
59
+ "loss": 2.8572198486328126,
60
+ "step": 600
61
+ },
62
+ {
63
+ "epoch": 0.2883625128733265,
64
+ "grad_norm": 0.43497762084007263,
65
+ "learning_rate": 0.003,
66
+ "loss": 2.7712442016601564,
67
+ "step": 700
68
+ },
69
+ {
70
+ "epoch": 0.329557157569516,
71
+ "grad_norm": 0.44306352734565735,
72
+ "learning_rate": 0.003,
73
+ "loss": 2.7086080932617187,
74
+ "step": 800
75
+ },
76
+ {
77
+ "epoch": 0.3707518022657055,
78
+ "grad_norm": 0.39232972264289856,
79
+ "learning_rate": 0.003,
80
+ "loss": 2.6619219970703125,
81
+ "step": 900
82
+ },
83
+ {
84
+ "epoch": 0.411946446961895,
85
+ "grad_norm": 0.34500154852867126,
86
+ "learning_rate": 0.003,
87
+ "loss": 2.6315206909179687,
88
+ "step": 1000
89
+ },
90
+ {
91
+ "epoch": 0.411946446961895,
92
+ "eval_loss": 2.648106813430786,
93
+ "eval_runtime": 7.4182,
94
+ "eval_samples_per_second": 432.722,
95
+ "eval_steps_per_second": 54.191,
96
+ "step": 1000
97
+ },
98
+ {
99
+ "epoch": 0.45314109165808447,
100
+ "grad_norm": 0.3188333213329315,
101
+ "learning_rate": 0.003,
102
+ "loss": 2.6055862426757814,
103
+ "step": 1100
104
+ },
105
+ {
106
+ "epoch": 0.49433573635427397,
107
+ "grad_norm": 0.29628098011016846,
108
+ "learning_rate": 0.003,
109
+ "loss": 2.5831158447265623,
110
+ "step": 1200
111
+ },
112
+ {
113
+ "epoch": 0.5355303810504635,
114
+ "grad_norm": 0.2794441282749176,
115
+ "learning_rate": 0.003,
116
+ "loss": 2.567415771484375,
117
+ "step": 1300
118
+ },
119
+ {
120
+ "epoch": 0.576725025746653,
121
+ "grad_norm": 0.2878789007663727,
122
+ "learning_rate": 0.003,
123
+ "loss": 2.5535826110839843,
124
+ "step": 1400
125
+ },
126
+ {
127
+ "epoch": 0.6179196704428425,
128
+ "grad_norm": 0.3040012717247009,
129
+ "learning_rate": 0.003,
130
+ "loss": 2.5389122009277343,
131
+ "step": 1500
132
+ },
133
+ {
134
+ "epoch": 0.6179196704428425,
135
+ "eval_loss": 2.558194637298584,
136
+ "eval_runtime": 8.0558,
137
+ "eval_samples_per_second": 398.469,
138
+ "eval_steps_per_second": 49.902,
139
+ "step": 1500
140
+ },
141
+ {
142
+ "epoch": 0.659114315139032,
143
+ "grad_norm": 0.2754426598548889,
144
+ "learning_rate": 0.003,
145
+ "loss": 2.5258233642578123,
146
+ "step": 1600
147
+ },
148
+ {
149
+ "epoch": 0.7003089598352215,
150
+ "grad_norm": 0.2642139196395874,
151
+ "learning_rate": 0.003,
152
+ "loss": 2.5123500061035156,
153
+ "step": 1700
154
+ },
155
+ {
156
+ "epoch": 0.741503604531411,
157
+ "grad_norm": 0.31137749552726746,
158
+ "learning_rate": 0.003,
159
+ "loss": 2.5001824951171874,
160
+ "step": 1800
161
+ },
162
+ {
163
+ "epoch": 0.7826982492276005,
164
+ "grad_norm": 0.2772506773471832,
165
+ "learning_rate": 0.003,
166
+ "loss": 2.4918637084960937,
167
+ "step": 1900
168
+ },
169
+ {
170
+ "epoch": 0.82389289392379,
171
+ "grad_norm": 0.26965758204460144,
172
+ "learning_rate": 0.002899325557098001,
173
+ "loss": 2.4782290649414063,
174
+ "step": 2000
175
+ },
176
+ {
177
+ "epoch": 0.82389289392379,
178
+ "eval_loss": 2.5082216262817383,
179
+ "eval_runtime": 8.1628,
180
+ "eval_samples_per_second": 393.246,
181
+ "eval_steps_per_second": 49.248,
182
+ "step": 2000
183
+ },
184
+ {
185
+ "epoch": 0.8650875386199794,
186
+ "grad_norm": 0.23774883151054382,
187
+ "learning_rate": 0.0022915870802995226,
188
+ "loss": 2.467481842041016,
189
+ "step": 2100
190
+ },
191
+ {
192
+ "epoch": 0.9062821833161689,
193
+ "grad_norm": 0.2045634537935257,
194
+ "learning_rate": 0.0013644373889485982,
195
+ "loss": 2.442911834716797,
196
+ "step": 2200
197
+ },
198
+ {
199
+ "epoch": 0.9474768280123584,
200
+ "grad_norm": 0.20750091969966888,
201
+ "learning_rate": 0.0004919882091667163,
202
+ "loss": 2.410100555419922,
203
+ "step": 2300
204
+ },
205
+ {
206
+ "epoch": 0.9886714727085479,
207
+ "grad_norm": 0.15250912308692932,
208
+ "learning_rate": 2.6279207952956684e-05,
209
+ "loss": 2.3908013916015625,
210
+ "step": 2400
211
+ },
212
+ {
213
+ "epoch": 1.0,
214
+ "eval_loss": 2.4129438400268555,
215
+ "eval_runtime": 8.274,
216
+ "eval_samples_per_second": 387.961,
217
+ "eval_steps_per_second": 48.586,
218
+ "step": 2428
219
+ }
220
+ ],
221
+ "logging_steps": 100,
222
+ "max_steps": 2428,
223
+ "num_input_tokens_seen": 0,
224
+ "num_train_epochs": 1,
225
+ "save_steps": 500,
226
+ "stateful_callbacks": {
227
+ "TrainerControl": {
228
+ "args": {
229
+ "should_epoch_stop": false,
230
+ "should_evaluate": false,
231
+ "should_log": false,
232
+ "should_save": true,
233
+ "should_training_stop": true
234
+ },
235
+ "attributes": {}
236
+ }
237
+ },
238
+ "total_flos": 4331665489920000.0,
239
+ "train_batch_size": 400,
240
+ "trial_name": null,
241
+ "trial_params": null
242
+ }