| attention_logit_softcapping: null |
| attention_scores_scalar: null |
| bias: false |
| block_size: 8192 |
| final_logit_softcapping: null |
| gelu_approximate: none |
| head_size: 128 |
| hf_config: |
| name: Meta-Llama-3.1-8B |
| org: meta-llama |
| intermediate_size: 14336 |
| lm_head_bias: false |
| mlp_class_name: LLaMAMLP |
| n_embd: 4096 |
| n_expert: 0 |
| n_expert_per_token: 0 |
| n_head: 32 |
| n_layer: 32 |
| n_query_groups: 8 |
| name: Llama-3.1-8B |
| norm_class_name: RMSNorm |
| norm_eps: 1.0e-05 |
| padded_vocab_size: 128256 |
| padding_multiple: 512 |
| parallel_residual: false |
| post_attention_norm: false |
| post_mlp_norm: false |
| rope_base: 500000 |
| rope_condense_ratio: 1 |
| rotary_percentage: 1.0 |
| scale_embeddings: false |
| shared_attention_norm: false |
| sliding_window_layer_placing: null |
| sliding_window_size: null |
| vocab_size: 128000 |
|
|
