config.json
795 B · 35 lines · json Raw
1 {
2 "architectures": [
3 "LlamaForScore"
4 ],
5 "attention_bias": false,
6 "attention_dropout": 0.0,
7 "bias": false,
8 "bos_token_id": 1,
9 "do_normalize": false,
10 "eos_token_id": 2,
11 "hidden_act": "silu",
12 "hidden_size": 4096,
13 "initializer_range": 0.02,
14 "intermediate_size": 11008,
15 "max_position_embeddings": 2048,
16 "model_type": "llama",
17 "normalizer_type": null,
18 "num_attention_heads": 32,
19 "num_hidden_layers": 32,
20 "num_key_value_heads": 32,
21 "pad_token_id": 32000,
22 "pretraining_tp": 1,
23 "rms_norm_eps": 1e-06,
24 "rope_scaling": null,
25 "rope_theta": 10000.0,
26 "score_bias": true,
27 "score_dim": 1,
28 "score_type": "reward",
29 "tie_word_embeddings": false,
30 "torch_dtype": "bfloat16",
31 "transformers_version": "4.37.2",
32 "use_cache": true,
33 "vocab_size": 32001
34 }
35