myBit-Llama2-jp-127M-3 / config.json
HachiML's picture
Training in progress, step 2000
86e06c0 verified
raw
history blame contribute delete
815 Bytes
{
"architectures": [
"BitLlamaForCausalLM"
],
"attention_bias": false,
"attention_dropout": 0.0,
"auto_map": {
"AutoConfig": "modeling_bit_llama.BitLlamaConfig",
"AutoModelForCausalLM": "modeling_bit_llama.BitLlamaForCausalLM"
},
"bits": 8,
"bos_token_id": 1,
"eos_token_id": 2,
"hidden_act": "silu",
"hidden_size": 768,
"initializer_range": 0.02,
"intermediate_size": 1536,
"max_position_embeddings": 1024,
"model_type": "bit_llama",
"n_ctx": 128,
"num_attention_heads": 12,
"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": "4.38.2",
"use_cache": true,
"vocab_size": 43176
}