config.json
883 B · 36 lines · json Raw
1 {
2 "_name_or_path": "NousResearch/Hermes-3-Llama-3.1-8B",
3 "architectures": [
4 "LlamaForCausalLM"
5 ],
6 "attention_bias": false,
7 "attention_dropout": 0.0,
8 "bos_token_id": 128000,
9 "eos_token_id": 128040,
10 "hidden_act": "silu",
11 "hidden_size": 4096,
12 "initializer_range": 0.02,
13 "intermediate_size": 14336,
14 "max_position_embeddings": 131072,
15 "mlp_bias": false,
16 "model_type": "llama",
17 "num_attention_heads": 32,
18 "num_hidden_layers": 32,
19 "num_key_value_heads": 8,
20 "pretraining_tp": 1,
21 "rms_norm_eps": 1e-05,
22 "rope_scaling": {
23 "factor": 8.0,
24 "high_freq_factor": 4.0,
25 "low_freq_factor": 1.0,
26 "original_max_position_embeddings": 8192,
27 "rope_type": "llama3"
28 },
29 "rope_theta": 500000.0,
30 "tie_word_embeddings": false,
31 "torch_dtype": "bfloat16",
32 "transformers_version": "4.44.0.dev0",
33 "use_cache": true,
34 "vocab_size": 128256
35 }
36