sn38-model-2018-1 / config.json
fitleech's picture
Upload model
4bb416a verified
Raw History Blame Contribute Delete
1.46 kB
{
"aux_gate_dim": 12,
"num_attention_heads": 14,
"pad_token_id": 32763,
"architectures": [
"NanoExpandForCausalLM"
],
"intermediate_size": 7168,
"rope_theta": 100000.0,
"final_logit_softcapping": 15.0,
"num_hidden_layers": 28,
"eos_token_id": 32763,
"num_key_value_heads": 14,
"max_position_embeddings": 2048,
"bos_token_id": 32759,
"layer_types": [
"sliding_attention",
"sliding_attention",
"sliding_attention",
"full_attention",
"sliding_attention",
"sliding_attention",
"sliding_attention",
"full_attention",
"sliding_attention",
"sliding_attention",
"sliding_attention",
"full_attention",
"sliding_attention",
"sliding_attention",
"sliding_attention",
"full_attention",
"sliding_attention",
"sliding_attention",
"sliding_attention",
"full_attention",
"sliding_attention",
"sliding_attention",
"sliding_attention",
"full_attention",
"sliding_attention",
"sliding_attention",
"sliding_attention",
"full_attention"
],
"attention_multiplier": 1.2,
"aux_layers": [
1,
3,
5,
7,
9,
11,
13,
15,
17,
19,
21,
23,
25,
27
],
"sliding_window": 512,
"vocab_size": 32768,
"hidden_size": 1792,
"tie_word_embeddings": false,
"tap_layer": 14,
"mix_gate_dim": 24,
"use_cache": false,
"model_type": "sn38-nanoexpand",
"torch_dtype": "bfloat16"
}