test / config.json
xiongjie's picture
Upload OPTForCausalLM
7ac9b69 verified
raw
history blame
1.35 kB
{
"_name_or_path": "facebook/opt-350m",
"_remove_final_layer_norm": false,
"activation_dropout": 0.0,
"activation_function": "relu",
"architectures": [
"OPTForCausalLM"
],
"attention_dropout": 0.0,
"bos_token_id": 2,
"do_layer_norm_before": false,
"dropout": 0.1,
"enable_bias": true,
"eos_token_id": 2,
"ffn_dim": 4096,
"hidden_size": 1024,
"init_std": 0.02,
"layer_norm_elementwise_affine": true,
"layerdrop": 0.0,
"max_position_embeddings": 2048,
"model_type": "opt",
"num_attention_heads": 16,
"num_hidden_layers": 24,
"pad_token_id": 1,
"prefix": "</s>",
"quantization_config": {
"batch_size": 1,
"bits": 2,
"block_name_to_quantize": null,
"cache_block_outputs": true,
"damp_percent": 0.1,
"dataset": "c4",
"desc_act": false,
"exllama_config": {
"version": 1
},
"group_size": 128,
"max_input_length": null,
"model_seqlen": null,
"module_name_preceding_first_block": null,
"modules_in_block_to_quantize": null,
"pad_token_id": null,
"quant_method": "gptq",
"sym": true,
"tokenizer": null,
"true_sequential": true,
"use_cuda_fp16": false,
"use_exllama": true
},
"torch_dtype": "float16",
"transformers_version": "4.37.2",
"use_cache": true,
"vocab_size": 50272,
"word_embed_proj_dim": 512
}