Upload folder using huggingface_hub
Browse files- smollm2-1p7b__gptq__wikitext2__n32__s1/BENCH_DONE +1 -0
- smollm2-1p7b__gptq__wikitext2__n32__s1/chat_template.jinja +6 -0
- smollm2-1p7b__gptq__wikitext2__n32__s1/config.json +90 -0
- smollm2-1p7b__gptq__wikitext2__n32__s1/generation_config.json +8 -0
- smollm2-1p7b__gptq__wikitext2__n32__s1/model.safetensors +3 -0
- smollm2-1p7b__gptq__wikitext2__n32__s1/quant_log.csv +169 -0
- smollm2-1p7b__gptq__wikitext2__n32__s1/quantize_config.json +47 -0
- smollm2-1p7b__gptq__wikitext2__n32__s1/tokenizer.json +0 -0
- smollm2-1p7b__gptq__wikitext2__n32__s1/tokenizer_config.json +19 -0
smollm2-1p7b__gptq__wikitext2__n32__s1/BENCH_DONE
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
2026-07-31T00:03:28.840404+00:00
|
smollm2-1p7b__gptq__wikitext2__n32__s1/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__s1/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_0o2cky07",
|
| 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__s1/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__s1/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:28d37d8b4216782922421dcc6c991c3c11c756f3f4df8b68b9afba452e123859
|
| 3 |
+
size 1040336952
|
smollm2-1p7b__gptq__wikitext2__n32__s1/quant_log.csv
ADDED
|
@@ -0,0 +1,169 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
layer,module,loss,samples,damp,time
|
| 2 |
+
0,self_attn.k_proj,0.0000140313,0.01000,2.757
|
| 3 |
+
0,self_attn.v_proj,0.0000005018,0.01000,2.759
|
| 4 |
+
0,self_attn.q_proj,0.0000190579,0.01000,2.769
|
| 5 |
+
0,self_attn.o_proj,0.0000003966,0.01000,0.515
|
| 6 |
+
0,mlp.gate_proj,0.0000208121,0.01000,1.473
|
| 7 |
+
0,mlp.up_proj,0.0000196140,0.01000,1.481
|
| 8 |
+
0,mlp.down_proj,0.0000090893,0.01000,2.226
|
| 9 |
+
1,self_attn.q_proj,0.0000532194,0.01000,2.704
|
| 10 |
+
1,self_attn.v_proj,0.0000114490,0.01000,2.728
|
| 11 |
+
1,self_attn.k_proj,0.0000475512,0.01000,2.731
|
| 12 |
+
1,self_attn.o_proj,0.0000045412,0.01000,0.510
|
| 13 |
+
1,mlp.up_proj,0.0000342806,0.01000,1.453
|
| 14 |
+
1,mlp.gate_proj,0.0000396502,0.01000,1.452
|
| 15 |
+
1,mlp.down_proj,0.0001545026,0.01000,2.223
|
| 16 |
+
2,self_attn.k_proj,0.0000787618,0.01000,2.711
|
| 17 |
+
2,self_attn.q_proj,0.0000877587,0.01000,2.726
|
| 18 |
+
2,self_attn.v_proj,0.0000308722,0.01000,2.728
|
| 19 |
+
2,self_attn.o_proj,0.0000116263,0.01000,0.514
|
| 20 |
+
2,mlp.gate_proj,0.0000616758,0.01000,1.493
|
| 21 |
+
2,mlp.up_proj,0.0000549021,0.01000,1.504
|
| 22 |
+
2,mlp.down_proj,0.0000402802,0.01000,2.250
|
| 23 |
+
3,self_attn.v_proj,0.0000453536,0.01000,2.779
|
| 24 |
+
3,self_attn.k_proj,0.0001079686,0.01000,2.789
|
| 25 |
+
3,self_attn.q_proj,0.0001075549,0.01000,2.801
|
| 26 |
+
3,self_attn.o_proj,0.0000147184,0.01000,0.512
|
| 27 |
+
3,mlp.gate_proj,0.0000774706,0.01000,1.582
|
| 28 |
+
3,mlp.up_proj,0.0000736327,0.01000,1.589
|
| 29 |
+
3,mlp.down_proj,0.0000662021,0.01000,2.222
|
| 30 |
+
4,self_attn.v_proj,0.0000511156,0.01000,2.760
|
| 31 |
+
4,self_attn.q_proj,0.0001204849,0.01000,2.766
|
| 32 |
+
4,self_attn.k_proj,0.0001167769,0.01000,2.770
|
| 33 |
+
4,self_attn.o_proj,0.0000249179,0.01000,0.516
|
| 34 |
+
4,mlp.up_proj,0.0000933222,0.01000,1.513
|
| 35 |
+
4,mlp.gate_proj,0.0000963994,0.01000,1.514
|
| 36 |
+
4,mlp.down_proj,0.0000950162,0.01000,2.228
|
| 37 |
+
5,self_attn.q_proj,0.0001165438,0.01000,2.665
|
| 38 |
+
5,self_attn.v_proj,0.0000519299,0.01000,2.675
|
| 39 |
+
5,self_attn.k_proj,0.0001125859,0.01000,2.682
|
| 40 |
+
5,self_attn.o_proj,0.0000320319,0.01000,0.517
|
| 41 |
+
5,mlp.gate_proj,0.0001143912,0.01000,1.454
|
| 42 |
+
5,mlp.up_proj,0.0001150300,0.01000,1.462
|
| 43 |
+
5,mlp.down_proj,0.0001235573,0.01000,2.231
|
| 44 |
+
6,self_attn.k_proj,0.0001296426,0.01000,2.708
|
| 45 |
+
6,self_attn.q_proj,0.0001377864,0.01000,2.710
|
| 46 |
+
6,self_attn.v_proj,0.0000640920,0.01000,2.712
|
| 47 |
+
6,self_attn.o_proj,0.0000418499,0.01000,0.508
|
| 48 |
+
6,mlp.gate_proj,0.0001349677,0.01000,1.432
|
| 49 |
+
6,mlp.up_proj,0.0001382114,0.01000,1.432
|
| 50 |
+
6,mlp.down_proj,0.0001650804,0.01000,2.218
|
| 51 |
+
7,self_attn.k_proj,0.0001335157,0.01000,2.619
|
| 52 |
+
7,self_attn.v_proj,0.0000657598,0.01000,2.628
|
| 53 |
+
7,self_attn.q_proj,0.0001645490,0.01000,2.636
|
| 54 |
+
7,self_attn.o_proj,0.0000646752,0.01000,0.515
|
| 55 |
+
7,mlp.gate_proj,0.0001584398,0.01000,1.632
|
| 56 |
+
7,mlp.up_proj,0.0001691150,0.01000,1.643
|
| 57 |
+
7,mlp.down_proj,0.0004878012,0.01000,2.220
|
| 58 |
+
8,self_attn.v_proj,0.0000886935,0.01000,2.835
|
| 59 |
+
8,self_attn.k_proj,0.0001272746,0.01000,2.861
|
| 60 |
+
8,self_attn.q_proj,0.0001287798,0.01000,2.874
|
| 61 |
+
8,self_attn.o_proj,0.0001136538,0.01000,0.530
|
| 62 |
+
8,mlp.up_proj,0.0001943407,0.01000,1.788
|
| 63 |
+
8,mlp.gate_proj,0.0001852683,0.01000,1.808
|
| 64 |
+
8,mlp.down_proj,0.0003005170,0.01000,2.234
|
| 65 |
+
9,self_attn.v_proj,0.0000764535,0.01000,2.803
|
| 66 |
+
9,self_attn.k_proj,0.0001187319,0.01000,2.817
|
| 67 |
+
9,self_attn.q_proj,0.0001192437,0.01000,2.824
|
| 68 |
+
9,self_attn.o_proj,0.0001573752,0.01000,0.516
|
| 69 |
+
9,mlp.gate_proj,0.0001768676,0.01000,1.531
|
| 70 |
+
9,mlp.up_proj,0.0002009571,0.01000,1.539
|
| 71 |
+
9,mlp.down_proj,0.0003198897,0.01000,2.250
|
| 72 |
+
10,self_attn.q_proj,0.0001290557,0.01000,2.816
|
| 73 |
+
10,self_attn.v_proj,0.0000790346,0.01000,2.816
|
| 74 |
+
10,self_attn.k_proj,0.0001264578,0.01000,2.828
|
| 75 |
+
10,self_attn.o_proj,0.0001946802,0.01000,0.512
|
| 76 |
+
10,mlp.up_proj,0.0001950986,0.01000,1.634
|
| 77 |
+
10,mlp.gate_proj,0.0001737697,0.01000,1.666
|
| 78 |
+
10,mlp.down_proj,0.0003717933,0.01000,2.224
|
| 79 |
+
11,self_attn.q_proj,0.0001299799,0.01000,2.872
|
| 80 |
+
11,self_attn.k_proj,0.0001255019,0.01000,2.874
|
| 81 |
+
11,self_attn.v_proj,0.0000773584,0.01000,2.883
|
| 82 |
+
11,self_attn.o_proj,0.0002108894,0.01000,0.512
|
| 83 |
+
11,mlp.up_proj,0.0001915785,0.01000,1.635
|
| 84 |
+
11,mlp.gate_proj,0.0001722518,0.01000,1.643
|
| 85 |
+
11,mlp.down_proj,0.0003934074,0.01000,2.243
|
| 86 |
+
12,self_attn.q_proj,0.0001465440,0.01000,2.796
|
| 87 |
+
12,self_attn.v_proj,0.0001104692,0.01000,2.819
|
| 88 |
+
12,self_attn.k_proj,0.0001377637,0.01000,2.829
|
| 89 |
+
12,self_attn.o_proj,0.0002722736,0.01000,0.515
|
| 90 |
+
12,mlp.gate_proj,0.0001795312,0.01000,1.523
|
| 91 |
+
12,mlp.up_proj,0.0002006539,0.01000,1.528
|
| 92 |
+
12,mlp.down_proj,0.0004303702,0.01000,2.241
|
| 93 |
+
13,self_attn.q_proj,0.0001396844,0.01000,2.732
|
| 94 |
+
13,self_attn.k_proj,0.0001340869,0.01000,2.732
|
| 95 |
+
13,self_attn.v_proj,0.0000971721,0.01000,2.737
|
| 96 |
+
13,self_attn.o_proj,0.0002283174,0.01000,0.512
|
| 97 |
+
13,mlp.gate_proj,0.0001836450,0.01000,1.572
|
| 98 |
+
13,mlp.up_proj,0.0002066669,0.01000,1.581
|
| 99 |
+
13,mlp.down_proj,0.0005037283,0.01000,2.234
|
| 100 |
+
14,self_attn.k_proj,0.0001336205,0.01000,2.759
|
| 101 |
+
14,self_attn.v_proj,0.0001048673,0.01000,2.778
|
| 102 |
+
14,self_attn.q_proj,0.0001322405,0.01000,2.792
|
| 103 |
+
14,self_attn.o_proj,0.0002995336,0.01000,0.513
|
| 104 |
+
14,mlp.up_proj,0.0002024275,0.01000,1.509
|
| 105 |
+
14,mlp.gate_proj,0.0001759266,0.01000,1.515
|
| 106 |
+
14,mlp.down_proj,0.0005963360,0.01000,2.218
|
| 107 |
+
15,self_attn.v_proj,0.0001607588,0.01000,2.818
|
| 108 |
+
15,self_attn.q_proj,0.0001454919,0.01000,2.830
|
| 109 |
+
15,self_attn.k_proj,0.0001407737,0.01000,2.835
|
| 110 |
+
15,self_attn.o_proj,0.0003328647,0.01000,0.514
|
| 111 |
+
15,mlp.gate_proj,0.0001891266,0.01000,1.603
|
| 112 |
+
15,mlp.up_proj,0.0002209504,0.01000,1.612
|
| 113 |
+
15,mlp.down_proj,0.0008630865,0.01000,2.241
|
| 114 |
+
16,self_attn.k_proj,0.0001425484,0.01000,2.728
|
| 115 |
+
16,self_attn.q_proj,0.0001437933,0.01000,2.729
|
| 116 |
+
16,self_attn.v_proj,0.0001441568,0.01000,2.741
|
| 117 |
+
16,self_attn.o_proj,0.0004064945,0.01000,0.510
|
| 118 |
+
16,mlp.gate_proj,0.0002304499,0.01000,1.517
|
| 119 |
+
16,mlp.up_proj,0.0002731381,0.01000,1.532
|
| 120 |
+
16,mlp.down_proj,0.0014025022,0.01000,2.226
|
| 121 |
+
17,self_attn.k_proj,0.0001404218,0.01000,2.917
|
| 122 |
+
17,self_attn.q_proj,0.0001418251,0.01000,2.932
|
| 123 |
+
17,self_attn.v_proj,0.0002017879,0.01000,2.935
|
| 124 |
+
17,self_attn.o_proj,0.0007359358,0.01000,0.522
|
| 125 |
+
17,mlp.gate_proj,0.0002861184,0.01000,1.567
|
| 126 |
+
17,mlp.up_proj,0.0003358395,0.01000,1.621
|
| 127 |
+
17,mlp.down_proj,0.0021867924,0.01000,2.240
|
| 128 |
+
18,self_attn.v_proj,0.0002428768,0.01000,2.840
|
| 129 |
+
18,self_attn.q_proj,0.0001498825,0.01000,2.860
|
| 130 |
+
18,self_attn.k_proj,0.0001434310,0.01000,2.867
|
| 131 |
+
18,self_attn.o_proj,0.0008369975,0.01000,0.513
|
| 132 |
+
18,mlp.up_proj,0.0004433132,0.01000,1.578
|
| 133 |
+
18,mlp.gate_proj,0.0003642228,0.01000,1.595
|
| 134 |
+
18,mlp.down_proj,0.0036796257,0.01000,2.235
|
| 135 |
+
19,self_attn.k_proj,0.0001560566,0.01000,2.896
|
| 136 |
+
19,self_attn.v_proj,0.0003460908,0.01000,2.912
|
| 137 |
+
19,self_attn.q_proj,0.0001652066,0.01000,2.927
|
| 138 |
+
19,self_attn.o_proj,0.0009389454,0.01000,0.517
|
| 139 |
+
19,mlp.gate_proj,0.0004294308,0.01000,1.548
|
| 140 |
+
19,mlp.up_proj,0.0005402375,0.01000,1.555
|
| 141 |
+
19,mlp.down_proj,0.0056531932,0.01000,2.241
|
| 142 |
+
20,self_attn.v_proj,0.0004096783,0.01000,2.862
|
| 143 |
+
20,self_attn.q_proj,0.0001647081,0.01000,2.875
|
| 144 |
+
20,self_attn.k_proj,0.0001544541,0.01000,2.885
|
| 145 |
+
20,self_attn.o_proj,0.0009919845,0.01000,0.512
|
| 146 |
+
20,mlp.gate_proj,0.0004815090,0.01000,1.529
|
| 147 |
+
20,mlp.up_proj,0.0006154163,0.01000,1.539
|
| 148 |
+
20,mlp.down_proj,0.0074356711,0.01000,2.231
|
| 149 |
+
21,self_attn.q_proj,0.0001827845,0.01000,2.844
|
| 150 |
+
21,self_attn.v_proj,0.0005285636,0.01000,2.852
|
| 151 |
+
21,self_attn.k_proj,0.0001753249,0.01000,2.854
|
| 152 |
+
21,self_attn.o_proj,0.0015608276,0.01000,0.513
|
| 153 |
+
21,mlp.up_proj,0.0007029230,0.01000,1.543
|
| 154 |
+
21,mlp.gate_proj,0.0005405610,0.01000,1.565
|
| 155 |
+
21,mlp.down_proj,0.0096381567,0.01000,2.237
|
| 156 |
+
22,self_attn.q_proj,0.0001885757,0.01000,2.751
|
| 157 |
+
22,self_attn.k_proj,0.0001788051,0.01000,2.764
|
| 158 |
+
22,self_attn.v_proj,0.0005454400,0.01000,2.765
|
| 159 |
+
22,self_attn.o_proj,0.0025493964,0.01000,0.513
|
| 160 |
+
22,mlp.up_proj,0.0007948312,0.01000,1.720
|
| 161 |
+
22,mlp.gate_proj,0.0006167706,0.01000,1.742
|
| 162 |
+
22,mlp.down_proj,0.0127694625,0.01000,2.241
|
| 163 |
+
23,self_attn.k_proj,0.0001698369,0.01000,2.802
|
| 164 |
+
23,self_attn.v_proj,0.0006908564,0.01000,2.804
|
| 165 |
+
23,self_attn.q_proj,0.0001646895,0.01000,2.815
|
| 166 |
+
23,self_attn.o_proj,0.0064120675,0.01000,0.511
|
| 167 |
+
23,mlp.up_proj,0.0008558327,0.01000,1.520
|
| 168 |
+
23,mlp.gate_proj,0.0007080542,0.01000,1.534
|
| 169 |
+
23,mlp.down_proj,0.0281582735,0.01000,2.246
|
smollm2-1p7b__gptq__wikitext2__n32__s1/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_0o2cky07",
|
| 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__s1/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
smollm2-1p7b__gptq__wikitext2__n32__s1/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 |
+
}
|