stan-hua's picture
GPTQ model for microsoft/Phi-3-small-8k-instruct: {'bits': 4, 'dataset': 'c4', 'group_size': 128, 'desc_act': False}
5d6c698 verified
{
"_name_or_path": "microsoft/Phi-3-small-8k-instruct",
"architectures": [
"Phi3SmallForCausalLM"
],
"attention_bias": false,
"attention_dropout_prob": 0.0,
"auto_map": {
"AutoConfig": "microsoft/Phi-3-small-8k-instruct--configuration_phi3_small.Phi3SmallConfig",
"AutoModelForCausalLM": "microsoft/Phi-3-small-8k-instruct--modeling_phi3_small.Phi3SmallForCausalLM",
"AutoTokenizer": "microsoft/Phi-3-small-8k-instruct--tokenization_phi3_small.Phi3SmallTokenizer"
},
"blocksparse_block_size": 64,
"blocksparse_homo_head_pattern": false,
"blocksparse_num_local_blocks": 16,
"blocksparse_triton_kernel_block_size": 64,
"blocksparse_vert_stride": 8,
"bos_token_id": 100257,
"dense_attention_every_n_layers": 2,
"dummy_token_indices": [
100256,
100258,
100259,
100260,
100264,
100265,
100267,
100268,
100269,
100270,
100271,
100272,
100273,
100274,
100275,
100276,
100277,
100278,
100279,
100280,
100281,
100282,
100283,
100284,
100285,
100286,
100287,
100288,
100289,
100290,
100291,
100292,
100293,
100294,
100295,
100296,
100297,
100298,
100299,
100300,
100301,
100302,
100303,
100304,
100305,
100306,
100307,
100308,
100309,
100310,
100311,
100312,
100313,
100314,
100315,
100316,
100317,
100318,
100319,
100320,
100321,
100322,
100323,
100324,
100325,
100326,
100327,
100328,
100329,
100330,
100331,
100332,
100333,
100334,
100335,
100336,
100337,
100338,
100339,
100340,
100341,
100342,
100343,
100344,
100345,
100346,
100347,
100348,
100349,
100350,
100351
],
"embedding_dropout_prob": 0.1,
"eos_token_id": 100257,
"ff_dim_multiplier": null,
"ff_intermediate_size": 14336,
"ffn_dropout_prob": 0.1,
"gegelu_limit": 20.0,
"gegelu_pad_to_256": true,
"hidden_act": "gegelu",
"hidden_size": 4096,
"initializer_range": 0.02,
"layer_norm_epsilon": 1e-05,
"max_position_embeddings": 8192,
"model_type": "phi3small",
"mup_attn_multiplier": 1.0,
"mup_embedding_multiplier": 10.0,
"mup_use_scaling": true,
"mup_width_multiplier": 8.0,
"num_attention_heads": 32,
"num_hidden_layers": 32,
"num_key_value_heads": 8,
"pad_sequence_to_multiple_of_64": true,
"quantization_config": {
"batch_size": 1,
"bits": 4,
"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
},
"reorder_and_upcast_attn": false,
"rope_embedding_base": 1000000,
"rope_position_scale": 1.0,
"rope_scaling": null,
"torch_dtype": "float16",
"transformers_version": "4.41.2",
"use_cache": true,
"vocab_size": 100352
}