|
{ |
|
"_name_or_path": "mistralai/Mistral-7B-v0.1", |
|
"architectures": [ |
|
"SparseMistralforCausalLM" |
|
], |
|
"attention_dropout": 0.0, |
|
"auto_map": { |
|
"AutoConfig": "sparsification_sftt.SparseMistralConfig", |
|
"AutoModelForCausalLM": "sparsification_sftt.SparseMistralforCausalLM" |
|
}, |
|
"bos_token_id": 1, |
|
"eos_token_id": 2, |
|
"hidden_act": "silu", |
|
"hidden_size": 4096, |
|
"initializer_range": 0.02, |
|
"intermediate_size": 14336, |
|
"max_position_embeddings": 32768, |
|
"model_type": "sparse_mistral", |
|
"num_attention_heads": 32, |
|
"num_hidden_layers": 32, |
|
"num_key_value_heads": 8, |
|
"rms_norm_eps": 1e-05, |
|
"rope_theta": 10000.0, |
|
"sliding_window": 4096, |
|
"thresholds": [ |
|
0.049147434532642365, |
|
0.0631895586848259, |
|
0.13340020179748535, |
|
0.12738214433193207, |
|
0.14944833517074585, |
|
0.1574724167585373, |
|
0.1675025075674057, |
|
0.17151454091072083, |
|
0.17953860759735107, |
|
0.17953860759735107, |
|
0.19157472252845764, |
|
0.20160481333732605, |
|
0.20561684668064117, |
|
0.2096288800239563, |
|
0.22367100417613983, |
|
0.23771312832832336, |
|
0.255767285823822, |
|
0.2678034007549286, |
|
0.26980942487716675, |
|
0.27382147312164307, |
|
0.27582746744155884, |
|
0.27582746744155884, |
|
0.27582746744155884, |
|
0.27582746744155884, |
|
0.27582746744155884, |
|
0.27582746744155884, |
|
0.277833491563797, |
|
0.277833491563797, |
|
0.36409226059913635, |
|
0.5466399192810059, |
|
0.9017050862312317, |
|
1.0 |
|
], |
|
"tie_word_embeddings": false, |
|
"torch_dtype": "bfloat16", |
|
"transformers_version": "4.36.2", |
|
"use_cache": false, |
|
"use_sparse_model": true, |
|
"use_sparse_predictor": false, |
|
"use_sparse_regularization": false, |
|
"vocab_size": 32000 |
|
} |
|
|