Upload folder using huggingface_hub
Browse files- smollm2-1p7b__gptq__wikitext2__n512__s2/BENCH_DONE +1 -0
- smollm2-1p7b__gptq__wikitext2__n512__s2/chat_template.jinja +6 -0
- smollm2-1p7b__gptq__wikitext2__n512__s2/config.json +90 -0
- smollm2-1p7b__gptq__wikitext2__n512__s2/generation_config.json +8 -0
- smollm2-1p7b__gptq__wikitext2__n512__s2/model.safetensors +3 -0
- smollm2-1p7b__gptq__wikitext2__n512__s2/quant_log.csv +169 -0
- smollm2-1p7b__gptq__wikitext2__n512__s2/quantize_config.json +47 -0
- smollm2-1p7b__gptq__wikitext2__n512__s2/tokenizer.json +0 -0
- smollm2-1p7b__gptq__wikitext2__n512__s2/tokenizer_config.json +19 -0
smollm2-1p7b__gptq__wikitext2__n512__s2/BENCH_DONE
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
2026-07-31T01:11:39.840633+00:00
|
smollm2-1p7b__gptq__wikitext2__n512__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__n512__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_87e5g0nu",
|
| 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__n512__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__n512__s2/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:16d4e9509a914145a9165df0207b6c398ac14129163bb052a5f39ab82fded800
|
| 3 |
+
size 1040336952
|
smollm2-1p7b__gptq__wikitext2__n512__s2/quant_log.csv
ADDED
|
@@ -0,0 +1,169 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
layer,module,loss,samples,damp,time
|
| 2 |
+
0,self_attn.v_proj,0.0000000333,0.01000,2.816
|
| 3 |
+
0,self_attn.q_proj,0.0000012811,0.01000,2.825
|
| 4 |
+
0,self_attn.k_proj,0.0000009377,0.01000,2.832
|
| 5 |
+
0,self_attn.o_proj,0.0000000279,0.01000,0.519
|
| 6 |
+
0,mlp.up_proj,0.0000013019,0.01000,1.557
|
| 7 |
+
0,mlp.gate_proj,0.0000013815,0.01000,1.562
|
| 8 |
+
0,mlp.down_proj,0.0000007724,0.01000,2.258
|
| 9 |
+
1,self_attn.v_proj,0.0000007529,0.01000,2.809
|
| 10 |
+
1,self_attn.k_proj,0.0000031446,0.01000,2.815
|
| 11 |
+
1,self_attn.q_proj,0.0000035152,0.01000,2.817
|
| 12 |
+
1,self_attn.o_proj,0.0000003155,0.01000,0.525
|
| 13 |
+
1,mlp.up_proj,0.0000022557,0.01000,1.537
|
| 14 |
+
1,mlp.gate_proj,0.0000026097,0.01000,1.539
|
| 15 |
+
1,mlp.down_proj,0.0000112013,0.01000,2.261
|
| 16 |
+
2,self_attn.q_proj,0.0000057708,0.01000,2.809
|
| 17 |
+
2,self_attn.v_proj,0.0000020155,0.01000,2.816
|
| 18 |
+
2,self_attn.k_proj,0.0000051569,0.01000,2.826
|
| 19 |
+
2,self_attn.o_proj,0.0000008138,0.01000,0.524
|
| 20 |
+
2,mlp.gate_proj,0.0000040314,0.01000,1.503
|
| 21 |
+
2,mlp.up_proj,0.0000035875,0.01000,1.511
|
| 22 |
+
2,mlp.down_proj,0.0000030761,0.01000,2.246
|
| 23 |
+
3,self_attn.q_proj,0.0000070932,0.01000,2.762
|
| 24 |
+
3,self_attn.v_proj,0.0000029725,0.01000,2.769
|
| 25 |
+
3,self_attn.k_proj,0.0000071055,0.01000,2.771
|
| 26 |
+
3,self_attn.o_proj,0.0000010246,0.01000,0.517
|
| 27 |
+
3,mlp.gate_proj,0.0000050922,0.01000,1.532
|
| 28 |
+
3,mlp.up_proj,0.0000048381,0.01000,1.538
|
| 29 |
+
3,mlp.down_proj,0.0000050021,0.01000,2.253
|
| 30 |
+
4,self_attn.k_proj,0.0000076992,0.01000,3.039
|
| 31 |
+
4,self_attn.q_proj,0.0000079792,0.01000,3.045
|
| 32 |
+
4,self_attn.v_proj,0.0000033628,0.01000,3.051
|
| 33 |
+
4,self_attn.o_proj,0.0000017614,0.01000,0.525
|
| 34 |
+
4,mlp.up_proj,0.0000061771,0.01000,1.615
|
| 35 |
+
4,mlp.gate_proj,0.0000063830,0.01000,1.627
|
| 36 |
+
4,mlp.down_proj,0.0000071341,0.01000,2.268
|
| 37 |
+
5,self_attn.q_proj,0.0000077574,0.01000,2.975
|
| 38 |
+
5,self_attn.k_proj,0.0000074873,0.01000,2.998
|
| 39 |
+
5,self_attn.v_proj,0.0000034382,0.01000,3.005
|
| 40 |
+
5,self_attn.o_proj,0.0000022410,0.01000,0.598
|
| 41 |
+
5,mlp.up_proj,0.0000075994,0.01000,1.534
|
| 42 |
+
5,mlp.gate_proj,0.0000075600,0.01000,1.542
|
| 43 |
+
5,mlp.down_proj,0.0000092125,0.01000,2.263
|
| 44 |
+
6,self_attn.q_proj,0.0000091664,0.01000,2.846
|
| 45 |
+
6,self_attn.k_proj,0.0000085923,0.01000,2.867
|
| 46 |
+
6,self_attn.v_proj,0.0000042419,0.01000,2.876
|
| 47 |
+
6,self_attn.o_proj,0.0000028034,0.01000,0.521
|
| 48 |
+
6,mlp.up_proj,0.0000091521,0.01000,1.662
|
| 49 |
+
6,mlp.gate_proj,0.0000089437,0.01000,1.667
|
| 50 |
+
6,mlp.down_proj,0.0000123253,0.01000,2.268
|
| 51 |
+
7,self_attn.v_proj,0.0000043363,0.01000,2.950
|
| 52 |
+
7,self_attn.k_proj,0.0000088446,0.01000,2.971
|
| 53 |
+
7,self_attn.q_proj,0.0000109863,0.01000,2.977
|
| 54 |
+
7,self_attn.o_proj,0.0000044925,0.01000,0.514
|
| 55 |
+
7,mlp.gate_proj,0.0000103722,0.01000,1.530
|
| 56 |
+
7,mlp.up_proj,0.0000110638,0.01000,1.537
|
| 57 |
+
7,mlp.down_proj,0.0000332371,0.01000,2.271
|
| 58 |
+
8,self_attn.q_proj,0.0000084329,0.01000,2.837
|
| 59 |
+
8,self_attn.k_proj,0.0000083392,0.01000,2.835
|
| 60 |
+
8,self_attn.v_proj,0.0000057962,0.01000,2.848
|
| 61 |
+
8,self_attn.o_proj,0.0000078580,0.01000,0.521
|
| 62 |
+
8,mlp.gate_proj,0.0000123636,0.01000,1.564
|
| 63 |
+
8,mlp.up_proj,0.0000129451,0.01000,1.572
|
| 64 |
+
8,mlp.down_proj,0.0000227164,0.01000,2.296
|
| 65 |
+
9,self_attn.q_proj,0.0000079129,0.01000,2.891
|
| 66 |
+
9,self_attn.v_proj,0.0000050649,0.01000,2.910
|
| 67 |
+
9,self_attn.k_proj,0.0000078825,0.01000,2.919
|
| 68 |
+
9,self_attn.o_proj,0.0000108470,0.01000,0.517
|
| 69 |
+
9,mlp.up_proj,0.0000132258,0.01000,1.511
|
| 70 |
+
9,mlp.gate_proj,0.0000116503,0.01000,1.512
|
| 71 |
+
9,mlp.down_proj,0.0000233046,0.01000,2.289
|
| 72 |
+
10,self_attn.k_proj,0.0000083208,0.01000,2.741
|
| 73 |
+
10,self_attn.q_proj,0.0000085153,0.01000,2.749
|
| 74 |
+
10,self_attn.v_proj,0.0000051896,0.01000,2.761
|
| 75 |
+
10,self_attn.o_proj,0.0000132367,0.01000,0.515
|
| 76 |
+
10,mlp.gate_proj,0.0000115032,0.01000,1.532
|
| 77 |
+
10,mlp.up_proj,0.0000128913,0.01000,1.531
|
| 78 |
+
10,mlp.down_proj,0.0000269980,0.01000,2.282
|
| 79 |
+
11,self_attn.q_proj,0.0000085876,0.01000,2.948
|
| 80 |
+
11,self_attn.k_proj,0.0000083093,0.01000,2.952
|
| 81 |
+
11,self_attn.v_proj,0.0000050980,0.01000,2.963
|
| 82 |
+
11,self_attn.o_proj,0.0000143379,0.01000,0.539
|
| 83 |
+
11,mlp.up_proj,0.0000127134,0.01000,1.649
|
| 84 |
+
11,mlp.gate_proj,0.0000114351,0.01000,1.661
|
| 85 |
+
11,mlp.down_proj,0.0000284588,0.01000,2.305
|
| 86 |
+
12,self_attn.q_proj,0.0000096726,0.01000,2.958
|
| 87 |
+
12,self_attn.v_proj,0.0000072929,0.01000,2.967
|
| 88 |
+
12,self_attn.k_proj,0.0000090924,0.01000,2.969
|
| 89 |
+
12,self_attn.o_proj,0.0000191464,0.01000,0.511
|
| 90 |
+
12,mlp.up_proj,0.0000133469,0.01000,1.556
|
| 91 |
+
12,mlp.gate_proj,0.0000119817,0.01000,1.564
|
| 92 |
+
12,mlp.down_proj,0.0000316646,0.01000,2.257
|
| 93 |
+
13,self_attn.k_proj,0.0000089491,0.01000,2.931
|
| 94 |
+
13,self_attn.v_proj,0.0000064649,0.01000,2.938
|
| 95 |
+
13,self_attn.q_proj,0.0000093323,0.01000,2.951
|
| 96 |
+
13,self_attn.o_proj,0.0000165875,0.01000,0.512
|
| 97 |
+
13,mlp.up_proj,0.0000137162,0.01000,1.554
|
| 98 |
+
13,mlp.gate_proj,0.0000122159,0.01000,1.563
|
| 99 |
+
13,mlp.down_proj,0.0000361139,0.01000,2.259
|
| 100 |
+
14,self_attn.q_proj,0.0000087610,0.01000,3.321
|
| 101 |
+
14,self_attn.v_proj,0.0000069354,0.01000,3.322
|
| 102 |
+
14,self_attn.k_proj,0.0000088415,0.01000,3.340
|
| 103 |
+
14,self_attn.o_proj,0.0000206170,0.01000,0.513
|
| 104 |
+
14,mlp.gate_proj,0.0000118199,0.01000,1.643
|
| 105 |
+
14,mlp.up_proj,0.0000135845,0.01000,1.662
|
| 106 |
+
14,mlp.down_proj,0.0000437735,0.01000,2.252
|
| 107 |
+
15,self_attn.k_proj,0.0000093529,0.01000,2.852
|
| 108 |
+
15,self_attn.v_proj,0.0000106334,0.01000,2.853
|
| 109 |
+
15,self_attn.q_proj,0.0000096354,0.01000,2.860
|
| 110 |
+
15,self_attn.o_proj,0.0000239700,0.01000,0.597
|
| 111 |
+
15,mlp.up_proj,0.0000147293,0.01000,1.584
|
| 112 |
+
15,mlp.gate_proj,0.0000126494,0.01000,1.591
|
| 113 |
+
15,mlp.down_proj,0.0000632347,0.01000,2.268
|
| 114 |
+
16,self_attn.v_proj,0.0000095744,0.01000,3.069
|
| 115 |
+
16,self_attn.q_proj,0.0000095326,0.01000,3.093
|
| 116 |
+
16,self_attn.k_proj,0.0000094968,0.01000,3.100
|
| 117 |
+
16,self_attn.o_proj,0.0000290412,0.01000,0.522
|
| 118 |
+
16,mlp.gate_proj,0.0000153931,0.01000,1.645
|
| 119 |
+
16,mlp.up_proj,0.0000181999,0.01000,1.655
|
| 120 |
+
16,mlp.down_proj,0.0001021158,0.01000,2.263
|
| 121 |
+
17,self_attn.k_proj,0.0000093579,0.01000,2.984
|
| 122 |
+
17,self_attn.q_proj,0.0000094195,0.01000,2.990
|
| 123 |
+
17,self_attn.v_proj,0.0000134769,0.01000,2.999
|
| 124 |
+
17,self_attn.o_proj,0.0000490915,0.01000,0.525
|
| 125 |
+
17,mlp.gate_proj,0.0000192805,0.01000,1.714
|
| 126 |
+
17,mlp.up_proj,0.0000226115,0.01000,1.736
|
| 127 |
+
17,mlp.down_proj,0.0001624909,0.01000,2.251
|
| 128 |
+
18,self_attn.k_proj,0.0000096549,0.01000,3.043
|
| 129 |
+
18,self_attn.q_proj,0.0000100877,0.01000,3.058
|
| 130 |
+
18,self_attn.v_proj,0.0000163757,0.01000,3.067
|
| 131 |
+
18,self_attn.o_proj,0.0000589167,0.01000,0.525
|
| 132 |
+
18,mlp.up_proj,0.0000300164,0.01000,1.823
|
| 133 |
+
18,mlp.gate_proj,0.0000246796,0.01000,1.833
|
| 134 |
+
18,mlp.down_proj,0.0002798562,0.01000,2.271
|
| 135 |
+
19,self_attn.k_proj,0.0000106232,0.01000,2.948
|
| 136 |
+
19,self_attn.v_proj,0.0000235840,0.01000,2.970
|
| 137 |
+
19,self_attn.q_proj,0.0000111949,0.01000,2.975
|
| 138 |
+
19,self_attn.o_proj,0.0000694785,0.01000,0.523
|
| 139 |
+
19,mlp.gate_proj,0.0000292226,0.01000,1.539
|
| 140 |
+
19,mlp.up_proj,0.0000366574,0.01000,1.547
|
| 141 |
+
19,mlp.down_proj,0.0004464538,0.01000,2.274
|
| 142 |
+
20,self_attn.v_proj,0.0000284118,0.01000,2.952
|
| 143 |
+
20,self_attn.q_proj,0.0000114143,0.01000,2.982
|
| 144 |
+
20,self_attn.k_proj,0.0000107031,0.01000,2.986
|
| 145 |
+
20,self_attn.o_proj,0.0000727793,0.01000,0.518
|
| 146 |
+
20,mlp.gate_proj,0.0000334764,0.01000,1.682
|
| 147 |
+
20,mlp.up_proj,0.0000426923,0.01000,1.691
|
| 148 |
+
20,mlp.down_proj,0.0006072192,0.01000,2.350
|
| 149 |
+
21,self_attn.v_proj,0.0000369773,0.01000,2.877
|
| 150 |
+
21,self_attn.q_proj,0.0000128012,0.01000,2.909
|
| 151 |
+
21,self_attn.k_proj,0.0000122959,0.01000,2.909
|
| 152 |
+
21,self_attn.o_proj,0.0001182307,0.01000,0.528
|
| 153 |
+
21,mlp.gate_proj,0.0000378574,0.01000,1.649
|
| 154 |
+
21,mlp.up_proj,0.0000491586,0.01000,1.677
|
| 155 |
+
21,mlp.down_proj,0.0007975217,0.01000,2.275
|
| 156 |
+
22,self_attn.q_proj,0.0000132414,0.01000,3.303
|
| 157 |
+
22,self_attn.k_proj,0.0000125570,0.01000,3.356
|
| 158 |
+
22,self_attn.v_proj,0.0000381319,0.01000,3.370
|
| 159 |
+
22,self_attn.o_proj,0.0001885067,0.01000,0.521
|
| 160 |
+
22,mlp.gate_proj,0.0000434872,0.01000,1.658
|
| 161 |
+
22,mlp.up_proj,0.0000561142,0.01000,1.670
|
| 162 |
+
22,mlp.down_proj,0.0010418120,0.01000,2.274
|
| 163 |
+
23,self_attn.k_proj,0.0000119905,0.01000,2.964
|
| 164 |
+
23,self_attn.v_proj,0.0000484822,0.01000,2.985
|
| 165 |
+
23,self_attn.q_proj,0.0000115985,0.01000,3.003
|
| 166 |
+
23,self_attn.o_proj,0.0004749729,0.01000,0.564
|
| 167 |
+
23,mlp.up_proj,0.0000598573,0.01000,1.599
|
| 168 |
+
23,mlp.gate_proj,0.0000495547,0.01000,1.609
|
| 169 |
+
23,mlp.down_proj,0.0022441864,0.01000,2.260
|
smollm2-1p7b__gptq__wikitext2__n512__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_87e5g0nu",
|
| 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__n512__s2/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
smollm2-1p7b__gptq__wikitext2__n512__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 |
+
}
|