File size: 1,420 Bytes
0a3aa28 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 |
{
"_name_or_path": "/models/falcon-three-10b",
"architectures": [
"LlamaForCausalLM"
],
"attention_bias": false,
"attention_dropout": 0.0,
"bos_token_id": 1,
"eos_token_id": 11,
"head_dim": 256,
"hidden_act": "silu",
"hidden_size": 3072,
"initializer_range": 0.02,
"intermediate_size": 23040,
"max_position_embeddings": 32768,
"mlp_bias": false,
"model_type": "llama",
"num_attention_heads": 12,
"num_hidden_layers": 40,
"num_key_value_heads": 4,
"pretraining_tp": 1,
"quantization_config": {
"amp": true,
"autoround_version": "0.4.2",
"batch_size": 8,
"bits": 4,
"data_type": "int",
"dataset": "NeelNanda/pile-10k",
"enable_minmax_tuning": true,
"enable_norm_bias_tuning": false,
"enable_quanted_input": true,
"gradient_accumulate_steps": 1,
"group_size": 128,
"iters": 1000,
"low_gpu_mem_usage": false,
"lr": 0.001,
"minmax_lr": 0.001,
"modules_to_not_convert": [
"lm_head"
],
"nsamples": 512,
"quant_method": "awq",
"scale_dtype": "torch.float16",
"seqlen": 2048,
"sym": true,
"to_quant_block_names": null,
"version": "gemm",
"zero_point": false
},
"rms_norm_eps": 1e-06,
"rope_scaling": null,
"rope_theta": 1000042,
"tie_word_embeddings": false,
"torch_dtype": "float16",
"transformers_version": "4.47.0",
"use_cache": true,
"vocab_size": 131072
}
|