|
{ |
|
"_name_or_path": "dunzhang/stella_en_400M_v5", |
|
"architectures": [ |
|
"NewModel" |
|
], |
|
"attention_probs_dropout_prob": 0.0, |
|
"auto_map": { |
|
"AutoConfig": "dunzhang/stella_en_400M_v5--configuration.NewConfig", |
|
"AutoModel": "dunzhang/stella_en_400M_v5--modeling.NewModel" |
|
}, |
|
"classifier_dropout": null, |
|
"hidden_act": "gelu", |
|
"hidden_dropout_prob": 0.1, |
|
"hidden_size": 1024, |
|
"initializer_range": 0.02, |
|
"intermediate_size": 4096, |
|
"layer_norm_eps": 1e-12, |
|
"layer_norm_type": "layer_norm", |
|
"logn_attention_clip1": false, |
|
"logn_attention_scale": false, |
|
"max_position_embeddings": 8192, |
|
"model_type": "new", |
|
"num_attention_heads": 16, |
|
"num_hidden_layers": 24, |
|
"pack_qkv": true, |
|
"pad_token_id": 0, |
|
"position_embedding_type": "rope", |
|
"rope_scaling": { |
|
"factor": 2.0, |
|
"type": "ntk" |
|
}, |
|
"rope_theta": 160000, |
|
"torch_dtype": "float32", |
|
"transformers_version": "4.45.2", |
|
"type_vocab_size": 2, |
|
"unpad_inputs": true, |
|
"use_memory_efficient_attention": true, |
|
"vocab_size": 30528 |
|
} |
|
|