melhoushi/gpt_cp_h2048_d40_gbs200_tpp20.0_lp0.8_linear_linear_reverse
Updated • 6
Mostafa Elhoushi
melhoushi
AI & ML interests
Make ML faster, smaller, smarter.
Recent Activity
upvoted a paper about 11 hours ago
Unlocking Lossless Speedups in LLMs via Discrete Diffusion authored a paper 2 days ago
Don't Drop Dropout: Optimizing Layer Sparsity for Efficient LLM Training and InferenceOrganizations
melhoushi/common_pile_h2048_d40_gbs200_tpp20.0_lp0.8_linear_linear_reverse
Updated • 4
melhoushi/Moonlight-16B-A3B-Instruct-transformers-5.13.0
Text Generation • 16B • Updated • 35
melhoushi/gpt_cp_h1152_d23_gbs76_tpp20.0_lp0.4_linear_linear_reverse
Updated • 5
melhoushi/common_pile_h1152_d23_gbs76_tpp20.0_lp0.4_linear_linear_reverse
Updated • 4
melhoushi/gpt_cp_h1152_d23_gbs76_tpp20.0_lp0.4_linear_null
Updated • 6
melhoushi/common_pile_h1152_d23_gbs76_tpp20.0_lp0.4_linear_null
Updated • 3
melhoushi/gpt_cp_h896_d17_gbs66_tpp20.0_lp0.4_linear_null
Updated • 4
melhoushi/common_pile_h896_d17_gbs66_tpp20.0_lp0.4_linear_null
Updated • 4
melhoushi/gpt_cp_h640_d13_gbs48_tpp20.0_lp0.4_linear_null
Updated • 7
melhoushi/common_pile_h640_d13_gbs48_tpp20.0_lp0.4_linear_null
Updated • 4
melhoushi/gpt_cp_h640_d13_gbs48_tpp20.0_lp0.4_linear_linear_reverse
Updated • 8
melhoushi/common_pile_h640_d13_gbs48_tpp20.0_lp0.4_linear_linear_reverse
Updated • 3
melhoushi/gpt_cp_h896_d17_gbs66_tpp20.0_lp0.2_linear_null
Updated • 4
melhoushi/gpt_cp_h896_d17_gbs66_tpp20.0_lp0.4_linear_linear_reverse
Updated • 5
melhoushi/common_pile_h896_d17_gbs66_tpp20.0_lp0.4_linear_linear_reverse
Updated • 6
melhoushi/common_pile_h896_d17_gbs66_tpp20.0_lp0.2_linear_null
Updated • 4
melhoushi/common_pile_h1152_d23_gbs76_tpp20.0_lp0.2_linear_null
Updated • 4
melhoushi/common_pile_h896_d17_gbs66_tpp20.0_lp0.0_null
Updated • 4
melhoushi/common_pile_h640_d13_gbs48_tpp20.0_lp0.2_linear_null
Updated • 4
melhoushi/common_pile_h640_d13_gbs48_tpp20.0_lp0.0_null
Updated • 8
melhoushi/common_pile_h1152_d23_gbs76_tpp20.0_lp0.0_null
Updated • 6
melhoushi/gpt_cp_h640_d13_gbs48_tpp20.0_lp0.2_linear_null
Updated • 5
melhoushi/gpt_cp_h1152_d23_gbs76_tpp20.0_lp0.2_linear_null
Updated • 4
melhoushi/gpt_cp_h896_d17_gbs66_tpp20.0_lp0.0_null
Updated • 4
melhoushi/gpt_cp_h640_d13_gbs48_tpp20.0_lp0.0_null
Updated • 5
melhoushi/gpt_cp_h1152_d23_gbs76_tpp20.0_lp0.0_null
Updated • 4
melhoushi/common_pile_h640_d13_gbs48_tpp20.0_lp0.1_uniform_null
Updated • 4
melhoushi/common_pile_h896_d17_gbs66_tpp20.0_lp0.2_uniform_null
Updated • 4
melhoushi/common_pile_h640_d13_gbs48_tpp20.0_lp0.2_uniform_null
Updated • 4