XibinBayesZhou commited on
Commit
4a49386
·
verified ·
1 Parent(s): d23ad14

upload all basics

Browse files
config.json ADDED
@@ -0,0 +1,67 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "/zhouxibin/models/huggingface/Evolla-10B-hf",
3
+ "aligner_attention_probs_dropout_prob": 0.1,
4
+ "aligner_enable_bias": true,
5
+ "aligner_ffn_mult": 4,
6
+ "aligner_num_add_layers": 8,
7
+ "architectures": [
8
+ "EvollaForProteinText2Text"
9
+ ],
10
+ "attention_bias": false,
11
+ "attention_dropout": 0.0,
12
+ "bos_token_id": 128000,
13
+ "eos_token_id": 128009,
14
+ "hidden_act": "silu",
15
+ "hidden_size": 4096,
16
+ "initializer_range": 0.02,
17
+ "intermediate_size": 14336,
18
+ "llm_config": {
19
+ "_attn_implementation_autoset": true,
20
+ "gradient_checkpointing": false,
21
+ "model_type": "evolla_llm",
22
+ "sequence_aligner_config": {
23
+ "protein_encoder_dim": 4096
24
+ }
25
+ },
26
+ "max_position_embeddings": 8192,
27
+ "mlp_bias": false,
28
+ "model_type": "EvollaModel",
29
+ "num_attention_heads": 32,
30
+ "num_hidden_layers": 32,
31
+ "num_key_value_heads": 8,
32
+ "pretraining_tp": 1,
33
+ "protein_attention_probs_dropout_prob": 0.1,
34
+ "protein_config": {
35
+ "model_type": "evolla_protein",
36
+ "resampler_config": {
37
+ "output_repr_dim": 4096,
38
+ "protein_repr_dim": 1280
39
+ }
40
+ },
41
+ "protein_emb_layer_norm_before": false,
42
+ "protein_hidden_dropout_prob": 0.1,
43
+ "protein_hidden_size": 1280,
44
+ "protein_intermediate_size": 5120,
45
+ "protein_layer_norm_eps": 1e-05,
46
+ "protein_mask_token_id": 4,
47
+ "protein_max_position_embeddings": 1026,
48
+ "protein_num_attention_heads": 20,
49
+ "protein_num_hidden_layers": 33,
50
+ "protein_pad_token_id": 1,
51
+ "protein_position_embedding_type": "rotary",
52
+ "protein_token_dropout": true,
53
+ "protein_vocab_size": 446,
54
+ "resampler_depth": 6,
55
+ "resampler_dim_head": 64,
56
+ "resampler_ff_mult": 4,
57
+ "resampler_heads": 8,
58
+ "resampler_num_latents": 64,
59
+ "rms_norm_eps": 1e-05,
60
+ "rope_scaling": null,
61
+ "rope_theta": 500000.0,
62
+ "tie_word_embeddings": false,
63
+ "torch_dtype": "float32",
64
+ "transformers_version": "4.49.0.dev0",
65
+ "use_cache": false,
66
+ "vocab_size": 128256
67
+ }
generation_config.json ADDED
@@ -0,0 +1,11 @@
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 128000,
4
+ "eos_token_id": 128009,
5
+ "do_sample": true,
6
+ "max_new_tokens": 512,
7
+ "temperature": 0.6,
8
+ "top_p": 0.9,
9
+ "transformers_version": "4.49.0.dev0",
10
+ "use_cache": false
11
+ }
pytorch_model.bin.index.json ADDED
The diff for this file is too large to render. See raw diff