774M-03_09_2024 / config.json
rusticluftig's picture
Upload model from https://huggingface.co/Jonathan18/net9_miner3/commit/923b2a589707c6317a1c5f71410a5ca81ec3d3d2
565b18d
{
"_name_or_path": "/workspace/pretraining/my_prerain/upload",
"architectures": [
"LlamaForCausalLM"
],
"attention_bias": false,
"bos_token_id": 1,
"eos_token_id": 2,
"hidden_act": "silu",
"hidden_size": 2048,
"initializer_range": 0.02,
"intermediate_size": 6112,
"max_position_embeddings": 1024,
"model_type": "llama",
"num_attention_heads": 32,
"num_hidden_layers": 12,
"num_key_value_heads": 4,
"pretraining_tp": 1,
"rms_norm_eps": 1e-05,
"rope_scaling": null,
"rope_theta": 10000.0,
"tie_word_embeddings": false,
"torch_dtype": "float32",
"transformers_version": "4.34.1",
"use_cache": true,
"vocab_size": 50257
}