MuPT-v0-4096-1.97B / config.json
XiaoY1's picture
Upload config.json with huggingface_hub
ff600ff verified
raw
history blame
696 Bytes
{
"architectures": [
"LlamaForCausalLM"
],
"attention_bias": false,
"attention_dropout": 0.0,
"bos_token_id": 2,
"eos_token_id": 3,
"hidden_act": "silu",
"hidden_size": 1536,
"initializer_range": 0.02,
"intermediate_size": 6144,
"max_position_embeddings": 4096,
"model_type": "llama",
"num_attention_heads": 24,
"num_hidden_layers": 48,
"num_key_value_heads": 24,
"pad_token_id": 0,
"pretraining_tp": 1,
"rms_norm_eps": 1e-05,
"rope_scaling": null,
"rope_theta": 10000.0,
"tie_word_embeddings": false,
"torch_dtype": "bfloat16",
"tokenizer_class": "MuPTTokenizer",
"transformers_version": "4.36.2",
"use_cache": true,
"vocab_size": 50000
}