30 lines
743 B
JSON
30 lines
743 B
JSON
|
|
{
|
||
|
|
"_name_or_path": "/fsx/arthur/ArthurZ/persimmon-8b-base",
|
||
|
|
"architectures": [
|
||
|
|
"PersimmonForCausalLM"
|
||
|
|
],
|
||
|
|
"bos_token_id": 71013,
|
||
|
|
"eos_token_id": 71013,
|
||
|
|
"hidden_act": "relu2",
|
||
|
|
"hidden_dropout": 0.6,
|
||
|
|
"hidden_size": 4096,
|
||
|
|
"initializer_range": 0.02,
|
||
|
|
"intermediate_size": 16384,
|
||
|
|
"layer_norm_eps": 1e-05,
|
||
|
|
"max_position_embeddings": 16384,
|
||
|
|
"model_type": "persimmon",
|
||
|
|
"num_attention_heads": 64,
|
||
|
|
"num_hidden_layers": 36,
|
||
|
|
"num_key_value_heads": 64,
|
||
|
|
"pretraining_tp": 1,
|
||
|
|
"qk_layernorm": true,
|
||
|
|
"rms_norm_eps": 1e-06,
|
||
|
|
"rope_scaling": null,
|
||
|
|
"rope_theta": 25000.0,
|
||
|
|
"tie_word_embeddings": false,
|
||
|
|
"torch_dtype": "bfloat16",
|
||
|
|
"transformers_version": "4.34.0.dev0",
|
||
|
|
"use_cache": true,
|
||
|
|
"vocab_size": 262144
|
||
|
|
}
|