Upload folder using huggingface_hub
Browse files- smollm2-1p7b__gptq__wikitext2__n32__s2/BENCH_DONE +1 -0
- smollm2-1p7b__gptq__wikitext2__n32__s2/chat_template.jinja +6 -0
- smollm2-1p7b__gptq__wikitext2__n32__s2/config.json +90 -0
- smollm2-1p7b__gptq__wikitext2__n32__s2/generation_config.json +8 -0
- smollm2-1p7b__gptq__wikitext2__n32__s2/model.safetensors +3 -0
- smollm2-1p7b__gptq__wikitext2__n32__s2/quant_log.csv +169 -0
- smollm2-1p7b__gptq__wikitext2__n32__s2/quantize_config.json +47 -0
- smollm2-1p7b__gptq__wikitext2__n32__s2/tokenizer.json +0 -0
- smollm2-1p7b__gptq__wikitext2__n32__s2/tokenizer_config.json +19 -0
smollm2-1p7b__gptq__wikitext2__n32__s2/BENCH_DONE
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
2026-07-31T00:08:28.662372+00:00
|
smollm2-1p7b__gptq__wikitext2__n32__s2/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__s2/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_f6wfcgn8",
|
| 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__s2/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__s2/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1b28a89557cf28570dfc6d8fbc37e573bb9e91458a062d1d878c4c1747e4dba4
|
| 3 |
+
size 1040336952
|
smollm2-1p7b__gptq__wikitext2__n32__s2/quant_log.csv
ADDED
|
@@ -0,0 +1,169 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
layer,module,loss,samples,damp,time
|
| 2 |
+
0,self_attn.q_proj,0.0000183738,0.01000,2.822
|
| 3 |
+
0,self_attn.k_proj,0.0000135467,0.01000,2.837
|
| 4 |
+
0,self_attn.v_proj,0.0000004907,0.01000,2.850
|
| 5 |
+
0,self_attn.o_proj,0.0000004192,0.01000,0.517
|
| 6 |
+
0,mlp.up_proj,0.0000188856,0.01000,1.495
|
| 7 |
+
0,mlp.gate_proj,0.0000200471,0.01000,1.511
|
| 8 |
+
0,mlp.down_proj,0.0000087102,0.01000,2.262
|
| 9 |
+
1,self_attn.v_proj,0.0000111660,0.01000,2.746
|
| 10 |
+
1,self_attn.q_proj,0.0000522674,0.01000,2.780
|
| 11 |
+
1,self_attn.k_proj,0.0000466868,0.01000,2.787
|
| 12 |
+
1,self_attn.o_proj,0.0000045803,0.01000,0.529
|
| 13 |
+
1,mlp.up_proj,0.0000332831,0.01000,1.558
|
| 14 |
+
1,mlp.gate_proj,0.0000384329,0.01000,1.587
|
| 15 |
+
1,mlp.down_proj,0.0001594285,0.01000,2.264
|
| 16 |
+
2,self_attn.k_proj,0.0000769328,0.01000,2.728
|
| 17 |
+
2,self_attn.q_proj,0.0000857287,0.01000,2.730
|
| 18 |
+
2,self_attn.v_proj,0.0000301183,0.01000,2.743
|
| 19 |
+
2,self_attn.o_proj,0.0000120961,0.01000,0.518
|
| 20 |
+
2,mlp.up_proj,0.0000539622,0.01000,1.522
|
| 21 |
+
2,mlp.gate_proj,0.0000605377,0.01000,1.527
|
| 22 |
+
2,mlp.down_proj,0.0000399642,0.01000,2.273
|
| 23 |
+
3,self_attn.q_proj,0.0001061481,0.01000,2.982
|
| 24 |
+
3,self_attn.k_proj,0.0001063853,0.01000,3.002
|
| 25 |
+
3,self_attn.v_proj,0.0000447491,0.01000,3.013
|
| 26 |
+
3,self_attn.o_proj,0.0000146135,0.01000,0.518
|
| 27 |
+
3,mlp.up_proj,0.0000721214,0.01000,1.556
|
| 28 |
+
3,mlp.gate_proj,0.0000758348,0.01000,1.571
|
| 29 |
+
3,mlp.down_proj,0.0000639821,0.01000,2.278
|
| 30 |
+
4,self_attn.v_proj,0.0000502024,0.01000,2.839
|
| 31 |
+
4,self_attn.k_proj,0.0001146505,0.01000,2.863
|
| 32 |
+
4,self_attn.q_proj,0.0001185110,0.01000,2.876
|
| 33 |
+
4,self_attn.o_proj,0.0000236339,0.01000,0.524
|
| 34 |
+
4,mlp.up_proj,0.0000924398,0.01000,1.541
|
| 35 |
+
4,mlp.gate_proj,0.0000953906,0.01000,1.567
|
| 36 |
+
4,mlp.down_proj,0.0000913894,0.01000,2.294
|
| 37 |
+
5,self_attn.q_proj,0.0001148734,0.01000,2.798
|
| 38 |
+
5,self_attn.v_proj,0.0000511927,0.01000,2.869
|
| 39 |
+
5,self_attn.k_proj,0.0001111858,0.01000,2.878
|
| 40 |
+
5,self_attn.o_proj,0.0000322084,0.01000,0.526
|
| 41 |
+
5,mlp.gate_proj,0.0001123384,0.01000,1.589
|
| 42 |
+
5,mlp.up_proj,0.0001129526,0.01000,1.597
|
| 43 |
+
5,mlp.down_proj,0.0001202117,0.01000,2.282
|
| 44 |
+
6,self_attn.k_proj,0.0001281009,0.01000,2.859
|
| 45 |
+
6,self_attn.v_proj,0.0000633522,0.01000,2.869
|
| 46 |
+
6,self_attn.q_proj,0.0001360824,0.01000,2.880
|
| 47 |
+
6,self_attn.o_proj,0.0000414110,0.01000,0.525
|
| 48 |
+
6,mlp.up_proj,0.0001380303,0.01000,1.570
|
| 49 |
+
6,mlp.gate_proj,0.0001345089,0.01000,1.586
|
| 50 |
+
6,mlp.down_proj,0.0001655682,0.01000,2.267
|
| 51 |
+
7,self_attn.v_proj,0.0000656146,0.01000,2.830
|
| 52 |
+
7,self_attn.k_proj,0.0001329507,0.01000,2.839
|
| 53 |
+
7,self_attn.q_proj,0.0001639505,0.01000,2.845
|
| 54 |
+
7,self_attn.o_proj,0.0000655735,0.01000,0.520
|
| 55 |
+
7,mlp.gate_proj,0.0001573663,0.01000,1.531
|
| 56 |
+
7,mlp.up_proj,0.0001680821,0.01000,1.547
|
| 57 |
+
7,mlp.down_proj,0.0004725568,0.01000,2.262
|
| 58 |
+
8,self_attn.v_proj,0.0000882143,0.01000,2.763
|
| 59 |
+
8,self_attn.q_proj,0.0001275864,0.01000,2.784
|
| 60 |
+
8,self_attn.k_proj,0.0001260603,0.01000,2.790
|
| 61 |
+
8,self_attn.o_proj,0.0001158254,0.01000,0.520
|
| 62 |
+
8,mlp.gate_proj,0.0001839297,0.01000,1.608
|
| 63 |
+
8,mlp.up_proj,0.0001928527,0.01000,1.611
|
| 64 |
+
8,mlp.down_proj,0.0003067816,0.01000,2.266
|
| 65 |
+
9,self_attn.k_proj,0.0001178260,0.01000,2.770
|
| 66 |
+
9,self_attn.v_proj,0.0000762973,0.01000,2.768
|
| 67 |
+
9,self_attn.q_proj,0.0001184957,0.01000,2.786
|
| 68 |
+
9,self_attn.o_proj,0.0001593639,0.01000,0.521
|
| 69 |
+
9,mlp.gate_proj,0.0001747827,0.01000,1.522
|
| 70 |
+
9,mlp.up_proj,0.0001985293,0.01000,1.542
|
| 71 |
+
9,mlp.down_proj,0.0003168676,0.01000,2.271
|
| 72 |
+
10,self_attn.v_proj,0.0000784652,0.01000,2.769
|
| 73 |
+
10,self_attn.k_proj,0.0001254408,0.01000,2.773
|
| 74 |
+
10,self_attn.q_proj,0.0001275455,0.01000,2.791
|
| 75 |
+
10,self_attn.o_proj,0.0001922510,0.01000,0.520
|
| 76 |
+
10,mlp.gate_proj,0.0001728801,0.01000,1.535
|
| 77 |
+
10,mlp.up_proj,0.0001940575,0.01000,1.540
|
| 78 |
+
10,mlp.down_proj,0.0003742078,0.01000,2.259
|
| 79 |
+
11,self_attn.q_proj,0.0001285961,0.01000,2.840
|
| 80 |
+
11,self_attn.k_proj,0.0001243517,0.01000,2.871
|
| 81 |
+
11,self_attn.v_proj,0.0000770105,0.01000,2.879
|
| 82 |
+
11,self_attn.o_proj,0.0002112082,0.01000,0.522
|
| 83 |
+
11,mlp.up_proj,0.0001938384,0.01000,1.587
|
| 84 |
+
11,mlp.gate_proj,0.0001741847,0.01000,1.605
|
| 85 |
+
11,mlp.down_proj,0.0003956165,0.01000,2.268
|
| 86 |
+
12,self_attn.q_proj,0.0001471435,0.01000,2.783
|
| 87 |
+
12,self_attn.v_proj,0.0001113918,0.01000,2.802
|
| 88 |
+
12,self_attn.k_proj,0.0001381362,0.01000,2.810
|
| 89 |
+
12,self_attn.o_proj,0.0002657627,0.01000,0.546
|
| 90 |
+
12,mlp.up_proj,0.0002017571,0.01000,1.974
|
| 91 |
+
12,mlp.gate_proj,0.0001808230,0.01000,2.003
|
| 92 |
+
12,mlp.down_proj,0.0004341918,0.01000,2.272
|
| 93 |
+
13,self_attn.q_proj,0.0001391974,0.01000,2.819
|
| 94 |
+
13,self_attn.v_proj,0.0000972509,0.01000,2.845
|
| 95 |
+
13,self_attn.k_proj,0.0001336211,0.01000,2.857
|
| 96 |
+
13,self_attn.o_proj,0.0002348000,0.01000,0.531
|
| 97 |
+
13,mlp.up_proj,0.0002074575,0.01000,1.619
|
| 98 |
+
13,mlp.gate_proj,0.0001843713,0.01000,1.638
|
| 99 |
+
13,mlp.down_proj,0.0005104564,0.01000,2.280
|
| 100 |
+
14,self_attn.q_proj,0.0001320775,0.01000,2.874
|
| 101 |
+
14,self_attn.v_proj,0.0001046866,0.01000,2.903
|
| 102 |
+
14,self_attn.k_proj,0.0001334746,0.01000,2.912
|
| 103 |
+
14,self_attn.o_proj,0.0002940003,0.01000,0.525
|
| 104 |
+
14,mlp.gate_proj,0.0001774185,0.01000,1.621
|
| 105 |
+
14,mlp.up_proj,0.0002044553,0.01000,1.629
|
| 106 |
+
14,mlp.down_proj,0.0006104992,0.01000,2.278
|
| 107 |
+
15,self_attn.q_proj,0.0001454084,0.01000,2.851
|
| 108 |
+
15,self_attn.k_proj,0.0001406278,0.01000,2.868
|
| 109 |
+
15,self_attn.v_proj,0.0001612254,0.01000,2.881
|
| 110 |
+
15,self_attn.o_proj,0.0003323500,0.01000,0.523
|
| 111 |
+
15,mlp.gate_proj,0.0001882133,0.01000,1.639
|
| 112 |
+
15,mlp.up_proj,0.0002198324,0.01000,1.650
|
| 113 |
+
15,mlp.down_proj,0.0008649052,0.01000,2.255
|
| 114 |
+
16,self_attn.k_proj,0.0001408727,0.01000,2.823
|
| 115 |
+
16,self_attn.q_proj,0.0001420253,0.01000,2.838
|
| 116 |
+
16,self_attn.v_proj,0.0001427209,0.01000,2.847
|
| 117 |
+
16,self_attn.o_proj,0.0004078785,0.01000,0.520
|
| 118 |
+
16,mlp.gate_proj,0.0002291718,0.01000,1.561
|
| 119 |
+
16,mlp.up_proj,0.0002716392,0.01000,1.578
|
| 120 |
+
16,mlp.down_proj,0.0014084384,0.01000,2.462
|
| 121 |
+
17,self_attn.v_proj,0.0001991923,0.01000,2.829
|
| 122 |
+
17,self_attn.k_proj,0.0001383540,0.01000,2.864
|
| 123 |
+
17,self_attn.q_proj,0.0001401035,0.01000,2.873
|
| 124 |
+
17,self_attn.o_proj,0.0006973090,0.01000,0.528
|
| 125 |
+
17,mlp.up_proj,0.0003355567,0.01000,1.615
|
| 126 |
+
17,mlp.gate_proj,0.0002858532,0.01000,1.629
|
| 127 |
+
17,mlp.down_proj,0.0022035902,0.01000,2.269
|
| 128 |
+
18,self_attn.k_proj,0.0001416869,0.01000,2.793
|
| 129 |
+
18,self_attn.q_proj,0.0001488134,0.01000,2.811
|
| 130 |
+
18,self_attn.v_proj,0.0002411216,0.01000,2.815
|
| 131 |
+
18,self_attn.o_proj,0.0008187315,0.01000,0.513
|
| 132 |
+
18,mlp.up_proj,0.0004420653,0.01000,1.617
|
| 133 |
+
18,mlp.gate_proj,0.0003630491,0.01000,1.634
|
| 134 |
+
18,mlp.down_proj,0.0036808751,0.01000,2.231
|
| 135 |
+
19,self_attn.v_proj,0.0003450972,0.01000,2.815
|
| 136 |
+
19,self_attn.q_proj,0.0001648123,0.01000,2.834
|
| 137 |
+
19,self_attn.k_proj,0.0001555229,0.01000,2.840
|
| 138 |
+
19,self_attn.o_proj,0.0009459148,0.01000,0.515
|
| 139 |
+
19,mlp.gate_proj,0.0004300785,0.01000,1.594
|
| 140 |
+
19,mlp.up_proj,0.0005410109,0.01000,1.607
|
| 141 |
+
19,mlp.down_proj,0.0056929262,0.01000,2.235
|
| 142 |
+
20,self_attn.v_proj,0.0004090156,0.01000,2.787
|
| 143 |
+
20,self_attn.k_proj,0.0001540709,0.01000,2.800
|
| 144 |
+
20,self_attn.q_proj,0.0001643993,0.01000,2.806
|
| 145 |
+
20,self_attn.o_proj,0.0009732789,0.01000,0.515
|
| 146 |
+
20,mlp.gate_proj,0.0004886559,0.01000,1.470
|
| 147 |
+
20,mlp.up_proj,0.0006238575,0.01000,1.474
|
| 148 |
+
20,mlp.down_proj,0.0076411506,0.01000,2.218
|
| 149 |
+
21,self_attn.q_proj,0.0001845382,0.01000,2.797
|
| 150 |
+
21,self_attn.v_proj,0.0005335110,0.01000,2.820
|
| 151 |
+
21,self_attn.k_proj,0.0001768411,0.01000,2.824
|
| 152 |
+
21,self_attn.o_proj,0.0015660517,0.01000,0.509
|
| 153 |
+
21,mlp.up_proj,0.0007148976,0.01000,1.498
|
| 154 |
+
21,mlp.gate_proj,0.0005498279,0.01000,1.502
|
| 155 |
+
21,mlp.down_proj,0.0099388808,0.01000,2.240
|
| 156 |
+
22,self_attn.k_proj,0.0001810904,0.01000,2.800
|
| 157 |
+
22,self_attn.q_proj,0.0001914714,0.01000,2.812
|
| 158 |
+
22,self_attn.v_proj,0.0005519100,0.01000,2.810
|
| 159 |
+
22,self_attn.o_proj,0.0025381714,0.01000,0.511
|
| 160 |
+
22,mlp.up_proj,0.0008058160,0.01000,1.500
|
| 161 |
+
22,mlp.gate_proj,0.0006251280,0.01000,1.507
|
| 162 |
+
22,mlp.down_proj,0.0133253373,0.01000,2.224
|
| 163 |
+
23,self_attn.k_proj,0.0001719539,0.01000,2.674
|
| 164 |
+
23,self_attn.v_proj,0.0007001865,0.01000,2.675
|
| 165 |
+
23,self_attn.q_proj,0.0001668679,0.01000,2.681
|
| 166 |
+
23,self_attn.o_proj,0.0066551911,0.01000,0.517
|
| 167 |
+
23,mlp.up_proj,0.0008581125,0.01000,1.514
|
| 168 |
+
23,mlp.gate_proj,0.0007098838,0.01000,1.522
|
| 169 |
+
23,mlp.down_proj,0.0283873826,0.01000,2.225
|
smollm2-1p7b__gptq__wikitext2__n32__s2/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_f6wfcgn8",
|
| 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__s2/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
smollm2-1p7b__gptq__wikitext2__n32__s2/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 |
+
}
|