appledora commited on
Commit
79217f6
·
verified ·
1 Parent(s): d51c60c

Upload ./recast8b_llama/config.json with huggingface_hub

Browse files
Files changed (1) hide show
  1. recast8b_llama/config.json +99 -0
recast8b_llama/config.json ADDED
@@ -0,0 +1,99 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "vocab_size": 128256,
3
+ "max_position_embeddings": 131072,
4
+ "hidden_size": 4096,
5
+ "intermediate_size": 14336,
6
+ "num_hidden_layers": 32,
7
+ "num_attention_heads": 32,
8
+ "num_key_value_heads": 8,
9
+ "hidden_act": "silu",
10
+ "initializer_range": 0.02,
11
+ "rms_norm_eps": 1e-05,
12
+ "pretraining_tp": 1,
13
+ "use_cache": true,
14
+ "mlp_bias": false,
15
+ "attention_bias": false,
16
+ "attention_dropout": 0.0,
17
+ "rope_theta": 500000.0,
18
+ "rope_scaling": {
19
+ "factor": 32.0,
20
+ "low_freq_factor": 1.0,
21
+ "high_freq_factor": 4.0,
22
+ "original_max_position_embeddings": 8192,
23
+ "rope_type": "llama3"
24
+ },
25
+ "torch_dtype": null,
26
+ "num_templates": 8,
27
+ "num_groups": 4,
28
+ "coef_height": 1,
29
+ "num_cf": 1,
30
+ "return_dict": true,
31
+ "output_hidden_states": false,
32
+ "output_attentions": false,
33
+ "torchscript": false,
34
+ "use_bfloat16": false,
35
+ "tf_legacy_loss": false,
36
+ "pruned_heads": {},
37
+ "tie_word_embeddings": false,
38
+ "chunk_size_feed_forward": 0,
39
+ "is_encoder_decoder": false,
40
+ "is_decoder": false,
41
+ "cross_attention_hidden_size": null,
42
+ "add_cross_attention": false,
43
+ "tie_encoder_decoder": false,
44
+ "max_length": 20,
45
+ "min_length": 0,
46
+ "do_sample": false,
47
+ "early_stopping": false,
48
+ "num_beams": 1,
49
+ "num_beam_groups": 1,
50
+ "diversity_penalty": 0.0,
51
+ "temperature": 1.0,
52
+ "top_k": 50,
53
+ "top_p": 1.0,
54
+ "typical_p": 1.0,
55
+ "repetition_penalty": 1.0,
56
+ "length_penalty": 1.0,
57
+ "no_repeat_ngram_size": 0,
58
+ "encoder_no_repeat_ngram_size": 0,
59
+ "bad_words_ids": null,
60
+ "num_return_sequences": 1,
61
+ "output_scores": false,
62
+ "return_dict_in_generate": false,
63
+ "forced_bos_token_id": null,
64
+ "forced_eos_token_id": null,
65
+ "remove_invalid_values": false,
66
+ "exponential_decay_length_penalty": null,
67
+ "suppress_tokens": null,
68
+ "begin_suppress_tokens": null,
69
+ "architectures": [
70
+ "RECAST8b_LlamaForCausalLM"
71
+ ],
72
+ "finetuning_task": null,
73
+ "id2label": {
74
+ "0": "LABEL_0",
75
+ "1": "LABEL_1"
76
+ },
77
+ "label2id": {
78
+ "LABEL_0": 0,
79
+ "LABEL_1": 1
80
+ },
81
+ "tokenizer_class": null,
82
+ "prefix": null,
83
+ "bos_token_id": 128000,
84
+ "pad_token_id": null,
85
+ "eos_token_id": 128001,
86
+ "sep_token_id": null,
87
+ "decoder_start_token_id": null,
88
+ "task_specific_params": null,
89
+ "problem_type": null,
90
+ "_name_or_path": "",
91
+ "transformers_version": "4.43.0.dev0",
92
+ "coef_width": 8,
93
+ "model_type": "recast8b_llama",
94
+ "auto_map": {
95
+ "AutoConfig": "configuration_recast_llama.RECAST8b_llama",
96
+ "AutoModel": "modeling_recast_llama.RECAST8b_llamaModel",
97
+ "AutoModelForCausalLM": "modeling_recast_llama.RECAST8b_LlamaForCausalLM"
98
+ }
99
+ }