phi-2-int4-inc / config.json
wenhuach's picture
replace with sym quantization as there is a large accuracy drop due to kernel issue
f201b74
raw
history blame
1.25 kB
{
"_name_or_path": "/models/phi-2",
"architectures": [
"PhiForCausalLM"
],
"attention_dropout": 0.0,
"bos_token_id": 50256,
"embd_pdrop": 0.0,
"eos_token_id": 50256,
"hidden_act": "gelu_new",
"hidden_size": 2560,
"initializer_range": 0.02,
"intermediate_size": 10240,
"layer_norm_eps": 1e-05,
"max_position_embeddings": 2048,
"model_type": "phi",
"num_attention_heads": 32,
"num_hidden_layers": 32,
"num_key_value_heads": 32,
"partial_rotary_factor": 0.4,
"qk_layernorm": false,
"quantization_config": {
"autoround_version": "0.2.0.dev",
"bits": 4,
"damp_percent": 0.01,
"desc_act": false,
"enable_minmax_tuning": true,
"enable_quanted_input": true,
"group_size": 128,
"is_marlin_format": false,
"iters": 1000,
"lr": 0.001,
"minmax_lr": 0.001,
"model_file_base_name": "model",
"model_name_or_path": null,
"quant_method": "gptq",
"scale_dtype": "float16",
"static_groups": false,
"sym": true,
"true_sequential": false
},
"resid_pdrop": 0.1,
"rope_scaling": null,
"rope_theta": 10000.0,
"tie_word_embeddings": false,
"torch_dtype": "float16",
"transformers_version": "4.40.2",
"use_cache": true,
"vocab_size": 51200
}