phi-3-orpo-v8_16 / config.json
cstr's picture
Trained with Unsloth
fcbf1ac verified
raw
history blame contribute delete
760 Bytes
{
"_name_or_path": "vonjack/Phi-3-mini-4k-instruct-LLaMAfied",
"architectures": [
"LlamaForCausalLM"
],
"attention_bias": false,
"attention_dropout": 0.0,
"bos_token_id": 1,
"eos_token_id": 32000,
"hidden_act": "silu",
"hidden_size": 3072,
"initializer_range": 0.02,
"intermediate_size": 8192,
"max_position_embeddings": 4096,
"model_type": "llama",
"num_attention_heads": 32,
"num_hidden_layers": 32,
"num_key_value_heads": 32,
"pretraining_tp": 1,
"rms_norm_eps": 1e-05,
"rope_scaling": null,
"rope_theta": 10000.0,
"sliding_window": 2047,
"tie_word_embeddings": false,
"torch_dtype": "float16",
"transformers_version": "4.40.0",
"unsloth_version": "2024.4",
"use_cache": true,
"vocab_size": 32064
}