Upload folder using huggingface_hub
Browse files- smollm2-1p7b__gptq__wikitext2__n32__s0/BENCH_DONE +1 -0
- smollm2-1p7b__gptq__wikitext2__n32__s0/chat_template.jinja +6 -0
- smollm2-1p7b__gptq__wikitext2__n32__s0/config.json +90 -0
- smollm2-1p7b__gptq__wikitext2__n32__s0/generation_config.json +8 -0
- smollm2-1p7b__gptq__wikitext2__n32__s0/model.safetensors +3 -0
- smollm2-1p7b__gptq__wikitext2__n32__s0/quant_log.csv +169 -0
- smollm2-1p7b__gptq__wikitext2__n32__s0/quantize_config.json +47 -0
- smollm2-1p7b__gptq__wikitext2__n32__s0/tokenizer.json +0 -0
- smollm2-1p7b__gptq__wikitext2__n32__s0/tokenizer_config.json +19 -0
smollm2-1p7b__gptq__wikitext2__n32__s0/BENCH_DONE
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
2026-07-30T23:57:05.492434+00:00
|
smollm2-1p7b__gptq__wikitext2__n32__s0/chat_template.jinja
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{% for message in messages %}{% if loop.first and messages[0]['role'] != 'system' %}{{ '<|im_start|>system
|
| 2 |
+
You are a helpful AI assistant named SmolLM, trained by Hugging Face<|im_end|>
|
| 3 |
+
' }}{% endif %}{{'<|im_start|>' + message['role'] + '
|
| 4 |
+
' + message['content'] + '<|im_end|>' + '
|
| 5 |
+
'}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant
|
| 6 |
+
' }}{% endif %}
|
smollm2-1p7b__gptq__wikitext2__n32__s0/config.json
ADDED
|
@@ -0,0 +1,90 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"LlamaForCausalLM"
|
| 4 |
+
],
|
| 5 |
+
"attention_bias": false,
|
| 6 |
+
"attention_dropout": 0.0,
|
| 7 |
+
"bos_token_id": 1,
|
| 8 |
+
"dtype": "bfloat16",
|
| 9 |
+
"eos_token_id": 2,
|
| 10 |
+
"head_dim": 64,
|
| 11 |
+
"hidden_act": "silu",
|
| 12 |
+
"hidden_size": 2048,
|
| 13 |
+
"initializer_range": 0.02,
|
| 14 |
+
"intermediate_size": 8192,
|
| 15 |
+
"max_position_embeddings": 8192,
|
| 16 |
+
"mlp_bias": false,
|
| 17 |
+
"model_type": "llama",
|
| 18 |
+
"num_attention_heads": 32,
|
| 19 |
+
"num_hidden_layers": 24,
|
| 20 |
+
"num_key_value_heads": 32,
|
| 21 |
+
"pad_token_id": 2,
|
| 22 |
+
"pretraining_tp": 1,
|
| 23 |
+
"quantization_config": {
|
| 24 |
+
"bits": 4,
|
| 25 |
+
"checkpoint_format": "gptq",
|
| 26 |
+
"desc_act": true,
|
| 27 |
+
"format": "gptq",
|
| 28 |
+
"group_size": 128,
|
| 29 |
+
"lm_head": false,
|
| 30 |
+
"meta": {
|
| 31 |
+
"act_group_aware": false,
|
| 32 |
+
"auto_forward_data_parallel": true,
|
| 33 |
+
"damp_auto_increment": 0.01,
|
| 34 |
+
"damp_percent": 0.01,
|
| 35 |
+
"dense_vram_strategy": "exclusive",
|
| 36 |
+
"dense_vram_strategy_devices": null,
|
| 37 |
+
"fallback": {
|
| 38 |
+
"smooth": null,
|
| 39 |
+
"strategy": "rtn",
|
| 40 |
+
"threshold": "0.5%"
|
| 41 |
+
},
|
| 42 |
+
"foem": null,
|
| 43 |
+
"gc_mode": "interval",
|
| 44 |
+
"gptaq": null,
|
| 45 |
+
"hessian": {
|
| 46 |
+
"chunk_bytes": null,
|
| 47 |
+
"chunk_size": null,
|
| 48 |
+
"staging_dtype": "float32"
|
| 49 |
+
},
|
| 50 |
+
"mock_quantization": false,
|
| 51 |
+
"moe_vram_strategy": "exclusive",
|
| 52 |
+
"moe_vram_strategy_devices": null,
|
| 53 |
+
"mse": 0.0,
|
| 54 |
+
"offload_to_disk": true,
|
| 55 |
+
"offload_to_disk_path": "/tmp/gptqmodel_jtsqg3q0",
|
| 56 |
+
"pack_impl": "cpu",
|
| 57 |
+
"quantizer": [
|
| 58 |
+
"gptqmodel:7.3.2"
|
| 59 |
+
],
|
| 60 |
+
"static_groups": false,
|
| 61 |
+
"true_sequential": true,
|
| 62 |
+
"uri": "https://github.com/modelcloud/gptqmodel",
|
| 63 |
+
"wait_for_submodule_finalizers": false
|
| 64 |
+
},
|
| 65 |
+
"method": "gptq",
|
| 66 |
+
"pack_dtype": "int32",
|
| 67 |
+
"quant_method": "gptq",
|
| 68 |
+
"sym": true
|
| 69 |
+
},
|
| 70 |
+
"rms_norm_eps": 1e-05,
|
| 71 |
+
"rope_parameters": {
|
| 72 |
+
"rope_theta": 130000,
|
| 73 |
+
"rope_type": "default"
|
| 74 |
+
},
|
| 75 |
+
"tie_word_embeddings": true,
|
| 76 |
+
"transformers.js_config": {
|
| 77 |
+
"dtype": "q4",
|
| 78 |
+
"kv_cache_dtype": {
|
| 79 |
+
"fp16": "float16",
|
| 80 |
+
"q4f16": "float16"
|
| 81 |
+
},
|
| 82 |
+
"use_external_data_format": {
|
| 83 |
+
"model.onnx": true,
|
| 84 |
+
"model_fp16.onnx": true
|
| 85 |
+
}
|
| 86 |
+
},
|
| 87 |
+
"transformers_version": "5.14.1",
|
| 88 |
+
"use_cache": true,
|
| 89 |
+
"vocab_size": 49152
|
| 90 |
+
}
|
smollm2-1p7b__gptq__wikitext2__n32__s0/generation_config.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 1,
|
| 4 |
+
"do_sample": true,
|
| 5 |
+
"eos_token_id": 2,
|
| 6 |
+
"pad_token_id": 2,
|
| 7 |
+
"transformers_version": "5.14.1"
|
| 8 |
+
}
|
smollm2-1p7b__gptq__wikitext2__n32__s0/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ebc7265c058349ffb77af7ac1e88d35615c3412bf7199e551d75b27a014332b0
|
| 3 |
+
size 1040336952
|
smollm2-1p7b__gptq__wikitext2__n32__s0/quant_log.csv
ADDED
|
@@ -0,0 +1,169 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
layer,module,loss,samples,damp,time
|
| 2 |
+
0,self_attn.v_proj,0.0000004896,0.01000,2.760
|
| 3 |
+
0,self_attn.k_proj,0.0000134341,0.01000,2.820
|
| 4 |
+
0,self_attn.q_proj,0.0000181882,0.01000,3.021
|
| 5 |
+
0,self_attn.o_proj,0.0000004098,0.01000,0.499
|
| 6 |
+
0,mlp.gate_proj,0.0000197816,0.01000,1.516
|
| 7 |
+
0,mlp.up_proj,0.0000186378,0.01000,1.536
|
| 8 |
+
0,mlp.down_proj,0.0000084761,0.01000,2.156
|
| 9 |
+
1,self_attn.q_proj,0.0000514789,0.01000,3.264
|
| 10 |
+
1,self_attn.v_proj,0.0000111260,0.01000,3.277
|
| 11 |
+
1,self_attn.k_proj,0.0000462241,0.01000,3.284
|
| 12 |
+
1,self_attn.o_proj,0.0000045965,0.01000,0.500
|
| 13 |
+
1,mlp.gate_proj,0.0000381194,0.01000,1.522
|
| 14 |
+
1,mlp.up_proj,0.0000329882,0.01000,1.535
|
| 15 |
+
1,mlp.down_proj,0.0001543988,0.01000,2.184
|
| 16 |
+
2,self_attn.k_proj,0.0000766402,0.01000,2.904
|
| 17 |
+
2,self_attn.v_proj,0.0000300597,0.01000,2.923
|
| 18 |
+
2,self_attn.q_proj,0.0000853173,0.01000,2.935
|
| 19 |
+
2,self_attn.o_proj,0.0000120826,0.01000,0.501
|
| 20 |
+
2,mlp.up_proj,0.0000534601,0.01000,1.585
|
| 21 |
+
2,mlp.gate_proj,0.0000600057,0.01000,1.597
|
| 22 |
+
2,mlp.down_proj,0.0000394039,0.01000,2.157
|
| 23 |
+
3,self_attn.k_proj,0.0001058467,0.01000,2.802
|
| 24 |
+
3,self_attn.v_proj,0.0000445078,0.01000,2.821
|
| 25 |
+
3,self_attn.q_proj,0.0001054465,0.01000,2.825
|
| 26 |
+
3,self_attn.o_proj,0.0000148973,0.01000,0.510
|
| 27 |
+
3,mlp.gate_proj,0.0000760561,0.01000,1.535
|
| 28 |
+
3,mlp.up_proj,0.0000723575,0.01000,1.555
|
| 29 |
+
3,mlp.down_proj,0.0000643225,0.01000,2.220
|
| 30 |
+
4,self_attn.q_proj,0.0001183025,0.01000,2.788
|
| 31 |
+
4,self_attn.k_proj,0.0001144139,0.01000,2.797
|
| 32 |
+
4,self_attn.v_proj,0.0000501840,0.01000,2.807
|
| 33 |
+
4,self_attn.o_proj,0.0000241545,0.01000,0.509
|
| 34 |
+
4,mlp.gate_proj,0.0000953587,0.01000,1.554
|
| 35 |
+
4,mlp.up_proj,0.0000923400,0.01000,1.566
|
| 36 |
+
4,mlp.down_proj,0.0000929211,0.01000,2.208
|
| 37 |
+
5,self_attn.k_proj,0.0001113959,0.01000,2.938
|
| 38 |
+
5,self_attn.q_proj,0.0001153068,0.01000,2.951
|
| 39 |
+
5,self_attn.v_proj,0.0000514457,0.01000,2.963
|
| 40 |
+
5,self_attn.o_proj,0.0000322932,0.01000,0.502
|
| 41 |
+
5,mlp.up_proj,0.0001138803,0.01000,1.541
|
| 42 |
+
5,mlp.gate_proj,0.0001131837,0.01000,1.555
|
| 43 |
+
5,mlp.down_proj,0.0001227695,0.01000,2.208
|
| 44 |
+
6,self_attn.k_proj,0.0001285943,0.01000,2.695
|
| 45 |
+
6,self_attn.v_proj,0.0000637237,0.01000,2.712
|
| 46 |
+
6,self_attn.q_proj,0.0001364791,0.01000,2.719
|
| 47 |
+
6,self_attn.o_proj,0.0000425618,0.01000,0.508
|
| 48 |
+
6,mlp.up_proj,0.0001382292,0.01000,1.502
|
| 49 |
+
6,mlp.gate_proj,0.0001350254,0.01000,1.509
|
| 50 |
+
6,mlp.down_proj,0.0001670711,0.01000,2.197
|
| 51 |
+
7,self_attn.v_proj,0.0000658106,0.01000,2.663
|
| 52 |
+
7,self_attn.q_proj,0.0001641770,0.01000,2.678
|
| 53 |
+
7,self_attn.k_proj,0.0001333539,0.01000,2.684
|
| 54 |
+
7,self_attn.o_proj,0.0000659343,0.01000,0.505
|
| 55 |
+
7,mlp.up_proj,0.0001695127,0.01000,1.561
|
| 56 |
+
7,mlp.gate_proj,0.0001588719,0.01000,1.558
|
| 57 |
+
7,mlp.down_proj,0.0005102395,0.01000,2.208
|
| 58 |
+
8,self_attn.q_proj,0.0001289410,0.01000,2.623
|
| 59 |
+
8,self_attn.v_proj,0.0000889823,0.01000,2.643
|
| 60 |
+
8,self_attn.k_proj,0.0001272693,0.01000,2.652
|
| 61 |
+
8,self_attn.o_proj,0.0001158078,0.01000,0.508
|
| 62 |
+
8,mlp.up_proj,0.0001975529,0.01000,1.429
|
| 63 |
+
8,mlp.gate_proj,0.0001884206,0.01000,1.440
|
| 64 |
+
8,mlp.down_proj,0.0003139318,0.01000,2.200
|
| 65 |
+
9,self_attn.v_proj,0.0000778240,0.01000,2.811
|
| 66 |
+
9,self_attn.k_proj,0.0001204078,0.01000,2.827
|
| 67 |
+
9,self_attn.q_proj,0.0001209218,0.01000,2.833
|
| 68 |
+
9,self_attn.o_proj,0.0001666145,0.01000,0.513
|
| 69 |
+
9,mlp.gate_proj,0.0001784506,0.01000,1.584
|
| 70 |
+
9,mlp.up_proj,0.0002027203,0.01000,1.590
|
| 71 |
+
9,mlp.down_proj,0.0003253785,0.01000,2.208
|
| 72 |
+
10,self_attn.v_proj,0.0000796959,0.01000,2.803
|
| 73 |
+
10,self_attn.k_proj,0.0001270298,0.01000,2.808
|
| 74 |
+
10,self_attn.q_proj,0.0001297231,0.01000,2.815
|
| 75 |
+
10,self_attn.o_proj,0.0001923665,0.01000,0.506
|
| 76 |
+
10,mlp.gate_proj,0.0001759876,0.01000,1.481
|
| 77 |
+
10,mlp.up_proj,0.0001974369,0.01000,1.483
|
| 78 |
+
10,mlp.down_proj,0.0003776983,0.01000,2.207
|
| 79 |
+
11,self_attn.k_proj,0.0001260549,0.01000,2.721
|
| 80 |
+
11,self_attn.v_proj,0.0000780243,0.01000,2.729
|
| 81 |
+
11,self_attn.q_proj,0.0001304119,0.01000,2.733
|
| 82 |
+
11,self_attn.o_proj,0.0002113247,0.01000,0.505
|
| 83 |
+
11,mlp.up_proj,0.0001944760,0.01000,1.466
|
| 84 |
+
11,mlp.gate_proj,0.0001748551,0.01000,1.478
|
| 85 |
+
11,mlp.down_proj,0.0003995131,0.01000,2.215
|
| 86 |
+
12,self_attn.v_proj,0.0001114354,0.01000,2.717
|
| 87 |
+
12,self_attn.k_proj,0.0001386506,0.01000,2.733
|
| 88 |
+
12,self_attn.q_proj,0.0001470835,0.01000,2.738
|
| 89 |
+
12,self_attn.o_proj,0.0002735543,0.01000,0.511
|
| 90 |
+
12,mlp.gate_proj,0.0001816626,0.01000,1.524
|
| 91 |
+
12,mlp.up_proj,0.0002028066,0.01000,1.531
|
| 92 |
+
12,mlp.down_proj,0.0004385323,0.01000,2.206
|
| 93 |
+
13,self_attn.k_proj,0.0001355494,0.01000,2.830
|
| 94 |
+
13,self_attn.v_proj,0.0000985314,0.01000,2.834
|
| 95 |
+
13,self_attn.q_proj,0.0001412900,0.01000,2.857
|
| 96 |
+
13,self_attn.o_proj,0.0002382883,0.01000,0.510
|
| 97 |
+
13,mlp.gate_proj,0.0001867163,0.01000,1.578
|
| 98 |
+
13,mlp.up_proj,0.0002098795,0.01000,1.584
|
| 99 |
+
13,mlp.down_proj,0.0005159752,0.01000,2.211
|
| 100 |
+
14,self_attn.v_proj,0.0001067252,0.01000,2.943
|
| 101 |
+
14,self_attn.k_proj,0.0001355535,0.01000,2.948
|
| 102 |
+
14,self_attn.q_proj,0.0001344807,0.01000,2.960
|
| 103 |
+
14,self_attn.o_proj,0.0003025114,0.01000,0.508
|
| 104 |
+
14,mlp.gate_proj,0.0001794669,0.01000,1.643
|
| 105 |
+
14,mlp.up_proj,0.0002067865,0.01000,1.645
|
| 106 |
+
14,mlp.down_proj,0.0006125814,0.01000,2.237
|
| 107 |
+
15,self_attn.k_proj,0.0001426496,0.01000,2.874
|
| 108 |
+
15,self_attn.q_proj,0.0001474779,0.01000,2.891
|
| 109 |
+
15,self_attn.v_proj,0.0001632364,0.01000,2.892
|
| 110 |
+
15,self_attn.o_proj,0.0003361382,0.01000,0.506
|
| 111 |
+
15,mlp.up_proj,0.0002248556,0.01000,1.579
|
| 112 |
+
15,mlp.gate_proj,0.0001926651,0.01000,1.591
|
| 113 |
+
15,mlp.down_proj,0.0008839607,0.01000,2.246
|
| 114 |
+
16,self_attn.k_proj,0.0001436103,0.01000,2.836
|
| 115 |
+
16,self_attn.v_proj,0.0001456100,0.01000,2.835
|
| 116 |
+
16,self_attn.q_proj,0.0001450439,0.01000,2.853
|
| 117 |
+
16,self_attn.o_proj,0.0004153942,0.01000,0.519
|
| 118 |
+
16,mlp.gate_proj,0.0002330936,0.01000,1.542
|
| 119 |
+
16,mlp.up_proj,0.0002760901,0.01000,1.552
|
| 120 |
+
16,mlp.down_proj,0.0014218048,0.01000,2.204
|
| 121 |
+
17,self_attn.q_proj,0.0001422536,0.01000,2.810
|
| 122 |
+
17,self_attn.v_proj,0.0002024632,0.01000,2.819
|
| 123 |
+
17,self_attn.k_proj,0.0001404730,0.01000,2.827
|
| 124 |
+
17,self_attn.o_proj,0.0007129266,0.01000,0.509
|
| 125 |
+
17,mlp.up_proj,0.0003392947,0.01000,1.425
|
| 126 |
+
17,mlp.gate_proj,0.0002890310,0.01000,1.426
|
| 127 |
+
17,mlp.down_proj,0.0022135614,0.01000,2.201
|
| 128 |
+
18,self_attn.v_proj,0.0002437818,0.01000,2.751
|
| 129 |
+
18,self_attn.k_proj,0.0001433552,0.01000,2.756
|
| 130 |
+
18,self_attn.q_proj,0.0001503983,0.01000,2.770
|
| 131 |
+
18,self_attn.o_proj,0.0008098470,0.01000,0.506
|
| 132 |
+
18,mlp.up_proj,0.0004497807,0.01000,1.500
|
| 133 |
+
18,mlp.gate_proj,0.0003692133,0.01000,1.508
|
| 134 |
+
18,mlp.down_proj,0.0037356950,0.01000,2.203
|
| 135 |
+
19,self_attn.q_proj,0.0001651468,0.01000,2.745
|
| 136 |
+
19,self_attn.k_proj,0.0001563465,0.01000,2.754
|
| 137 |
+
19,self_attn.v_proj,0.0003474685,0.01000,2.760
|
| 138 |
+
19,self_attn.o_proj,0.0009577784,0.01000,0.508
|
| 139 |
+
19,mlp.up_proj,0.0005402417,0.01000,1.564
|
| 140 |
+
19,mlp.gate_proj,0.0004292731,0.01000,1.575
|
| 141 |
+
19,mlp.down_proj,0.0057191895,0.01000,2.220
|
| 142 |
+
20,self_attn.k_proj,0.0001537593,0.01000,2.783
|
| 143 |
+
20,self_attn.v_proj,0.0004069071,0.01000,2.798
|
| 144 |
+
20,self_attn.q_proj,0.0001643704,0.01000,2.803
|
| 145 |
+
20,self_attn.o_proj,0.0009892953,0.01000,0.511
|
| 146 |
+
20,mlp.up_proj,0.0006214835,0.01000,1.569
|
| 147 |
+
20,mlp.gate_proj,0.0004862978,0.01000,1.571
|
| 148 |
+
20,mlp.down_proj,0.0076549966,0.01000,2.208
|
| 149 |
+
21,self_attn.v_proj,0.0005299663,0.01000,2.683
|
| 150 |
+
21,self_attn.q_proj,0.0001837524,0.01000,2.681
|
| 151 |
+
21,self_attn.k_proj,0.0001758970,0.01000,2.695
|
| 152 |
+
21,self_attn.o_proj,0.0015429526,0.01000,0.510
|
| 153 |
+
21,mlp.gate_proj,0.0005430855,0.01000,1.482
|
| 154 |
+
21,mlp.up_proj,0.0007060195,0.01000,1.487
|
| 155 |
+
21,mlp.down_proj,0.0098808324,0.01000,2.205
|
| 156 |
+
22,self_attn.q_proj,0.0001896824,0.01000,2.610
|
| 157 |
+
22,self_attn.k_proj,0.0001795994,0.01000,2.618
|
| 158 |
+
22,self_attn.v_proj,0.0005479072,0.01000,2.625
|
| 159 |
+
22,self_attn.o_proj,0.0025106594,0.01000,0.505
|
| 160 |
+
22,mlp.gate_proj,0.0006225423,0.01000,1.462
|
| 161 |
+
22,mlp.up_proj,0.0008025953,0.01000,1.467
|
| 162 |
+
22,mlp.down_proj,0.0131192869,0.01000,2.199
|
| 163 |
+
23,self_attn.v_proj,0.0006988762,0.01000,2.599
|
| 164 |
+
23,self_attn.k_proj,0.0001717117,0.01000,2.605
|
| 165 |
+
23,self_attn.q_proj,0.0001661855,0.01000,2.617
|
| 166 |
+
23,self_attn.o_proj,0.0064919991,0.01000,0.503
|
| 167 |
+
23,mlp.gate_proj,0.0007085501,0.01000,1.533
|
| 168 |
+
23,mlp.up_proj,0.0008565183,0.01000,1.541
|
| 169 |
+
23,mlp.down_proj,0.0281702951,0.01000,2.199
|
smollm2-1p7b__gptq__wikitext2__n32__s0/quantize_config.json
ADDED
|
@@ -0,0 +1,47 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bits": 4,
|
| 3 |
+
"group_size": 128,
|
| 4 |
+
"desc_act": true,
|
| 5 |
+
"lm_head": false,
|
| 6 |
+
"method": "gptq",
|
| 7 |
+
"quant_method": "gptq",
|
| 8 |
+
"format": "gptq",
|
| 9 |
+
"checkpoint_format": "gptq",
|
| 10 |
+
"pack_dtype": "int32",
|
| 11 |
+
"meta": {
|
| 12 |
+
"quantizer": [
|
| 13 |
+
"gptqmodel:7.3.2"
|
| 14 |
+
],
|
| 15 |
+
"uri": "https://github.com/modelcloud/gptqmodel",
|
| 16 |
+
"damp_percent": 0.01,
|
| 17 |
+
"damp_auto_increment": 0.01,
|
| 18 |
+
"static_groups": false,
|
| 19 |
+
"true_sequential": true,
|
| 20 |
+
"mse": 0.0,
|
| 21 |
+
"gptaq": null,
|
| 22 |
+
"foem": null,
|
| 23 |
+
"act_group_aware": false,
|
| 24 |
+
"fallback": {
|
| 25 |
+
"strategy": "rtn",
|
| 26 |
+
"threshold": "0.5%",
|
| 27 |
+
"smooth": null
|
| 28 |
+
},
|
| 29 |
+
"offload_to_disk": true,
|
| 30 |
+
"offload_to_disk_path": "/tmp/gptqmodel_jtsqg3q0",
|
| 31 |
+
"pack_impl": "cpu",
|
| 32 |
+
"gc_mode": "interval",
|
| 33 |
+
"wait_for_submodule_finalizers": false,
|
| 34 |
+
"auto_forward_data_parallel": true,
|
| 35 |
+
"dense_vram_strategy": "exclusive",
|
| 36 |
+
"dense_vram_strategy_devices": null,
|
| 37 |
+
"moe_vram_strategy": "exclusive",
|
| 38 |
+
"moe_vram_strategy_devices": null,
|
| 39 |
+
"mock_quantization": false,
|
| 40 |
+
"hessian": {
|
| 41 |
+
"chunk_size": null,
|
| 42 |
+
"chunk_bytes": null,
|
| 43 |
+
"staging_dtype": "float32"
|
| 44 |
+
}
|
| 45 |
+
},
|
| 46 |
+
"sym": true
|
| 47 |
+
}
|
smollm2-1p7b__gptq__wikitext2__n32__s0/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
smollm2-1p7b__gptq__wikitext2__n32__s0/tokenizer_config.json
ADDED
|
@@ -0,0 +1,19 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"backend": "tokenizers",
|
| 4 |
+
"bos_token": "<|im_start|>",
|
| 5 |
+
"clean_up_tokenization_spaces": false,
|
| 6 |
+
"eos_token": "<|im_end|>",
|
| 7 |
+
"errors": "replace",
|
| 8 |
+
"extra_special_tokens": [
|
| 9 |
+
"<|im_start|>",
|
| 10 |
+
"<|im_end|>"
|
| 11 |
+
],
|
| 12 |
+
"is_local": false,
|
| 13 |
+
"local_files_only": false,
|
| 14 |
+
"model_max_length": 8192,
|
| 15 |
+
"pad_token": "<|im_end|>",
|
| 16 |
+
"tokenizer_class": "GPT2Tokenizer",
|
| 17 |
+
"unk_token": "<|endoftext|>",
|
| 18 |
+
"vocab_size": 49152
|
| 19 |
+
}
|