Upload folder using huggingface_hub
Browse files- smollm2-1p7b__gptq__wikitext2__n512__s0/BENCH_DONE +1 -0
- smollm2-1p7b__gptq__wikitext2__n512__s0/chat_template.jinja +6 -0
- smollm2-1p7b__gptq__wikitext2__n512__s0/config.json +90 -0
- smollm2-1p7b__gptq__wikitext2__n512__s0/generation_config.json +8 -0
- smollm2-1p7b__gptq__wikitext2__n512__s0/model.safetensors +3 -0
- smollm2-1p7b__gptq__wikitext2__n512__s0/quant_log.csv +169 -0
- smollm2-1p7b__gptq__wikitext2__n512__s0/quantize_config.json +47 -0
- smollm2-1p7b__gptq__wikitext2__n512__s0/tokenizer.json +0 -0
- smollm2-1p7b__gptq__wikitext2__n512__s0/tokenizer_config.json +19 -0
smollm2-1p7b__gptq__wikitext2__n512__s0/BENCH_DONE
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
2026-07-31T00:43:07.269386+00:00
|
smollm2-1p7b__gptq__wikitext2__n512__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__n512__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_mjvln5mc",
|
| 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__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__n512__s0/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ccd36e69874608bd3f260d8fbcf730ee2a7bfc1cabf163f18558d237d5a3949a
|
| 3 |
+
size 1040336952
|
smollm2-1p7b__gptq__wikitext2__n512__s0/quant_log.csv
ADDED
|
@@ -0,0 +1,169 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
layer,module,loss,samples,damp,time
|
| 2 |
+
0,self_attn.q_proj,0.0000012826,0.01000,2.739
|
| 3 |
+
0,self_attn.v_proj,0.0000000332,0.01000,2.736
|
| 4 |
+
0,self_attn.k_proj,0.0000009407,0.01000,2.742
|
| 5 |
+
0,self_attn.o_proj,0.0000000278,0.01000,0.519
|
| 6 |
+
0,mlp.gate_proj,0.0000013832,0.01000,1.499
|
| 7 |
+
0,mlp.up_proj,0.0000013032,0.01000,1.508
|
| 8 |
+
0,mlp.down_proj,0.0000007734,0.01000,2.254
|
| 9 |
+
1,self_attn.k_proj,0.0000031449,0.01000,2.758
|
| 10 |
+
1,self_attn.v_proj,0.0000007533,0.01000,2.760
|
| 11 |
+
1,self_attn.q_proj,0.0000035185,0.01000,2.769
|
| 12 |
+
1,self_attn.o_proj,0.0000003180,0.01000,0.520
|
| 13 |
+
1,mlp.gate_proj,0.0000026163,0.01000,1.547
|
| 14 |
+
1,mlp.up_proj,0.0000022621,0.01000,1.546
|
| 15 |
+
1,mlp.down_proj,0.0000112607,0.01000,2.252
|
| 16 |
+
2,self_attn.k_proj,0.0000051742,0.01000,2.834
|
| 17 |
+
2,self_attn.q_proj,0.0000057852,0.01000,2.844
|
| 18 |
+
2,self_attn.v_proj,0.0000020217,0.01000,2.852
|
| 19 |
+
2,self_attn.o_proj,0.0000008231,0.01000,0.516
|
| 20 |
+
2,mlp.up_proj,0.0000036009,0.01000,1.629
|
| 21 |
+
2,mlp.gate_proj,0.0000040444,0.01000,1.627
|
| 22 |
+
2,mlp.down_proj,0.0000030936,0.01000,2.253
|
| 23 |
+
3,self_attn.q_proj,0.0000071391,0.01000,2.704
|
| 24 |
+
3,self_attn.k_proj,0.0000071677,0.01000,2.710
|
| 25 |
+
3,self_attn.v_proj,0.0000029915,0.01000,2.720
|
| 26 |
+
3,self_attn.o_proj,0.0000010234,0.01000,0.519
|
| 27 |
+
3,mlp.gate_proj,0.0000051600,0.01000,1.548
|
| 28 |
+
3,mlp.up_proj,0.0000048989,0.01000,1.558
|
| 29 |
+
3,mlp.down_proj,0.0000051378,0.01000,2.246
|
| 30 |
+
4,self_attn.v_proj,0.0000033989,0.01000,2.737
|
| 31 |
+
4,self_attn.k_proj,0.0000077856,0.01000,2.758
|
| 32 |
+
4,self_attn.q_proj,0.0000080543,0.01000,2.764
|
| 33 |
+
4,self_attn.o_proj,0.0000018202,0.01000,0.523
|
| 34 |
+
4,mlp.gate_proj,0.0000064159,0.01000,1.566
|
| 35 |
+
4,mlp.up_proj,0.0000062118,0.01000,1.565
|
| 36 |
+
4,mlp.down_proj,0.0000071873,0.01000,2.268
|
| 37 |
+
5,self_attn.q_proj,0.0000077669,0.01000,2.836
|
| 38 |
+
5,self_attn.k_proj,0.0000074912,0.01000,2.846
|
| 39 |
+
5,self_attn.v_proj,0.0000034477,0.01000,2.850
|
| 40 |
+
5,self_attn.o_proj,0.0000022289,0.01000,0.515
|
| 41 |
+
5,mlp.up_proj,0.0000076463,0.01000,1.655
|
| 42 |
+
5,mlp.gate_proj,0.0000076067,0.01000,1.666
|
| 43 |
+
5,mlp.down_proj,0.0000092938,0.01000,2.244
|
| 44 |
+
6,self_attn.k_proj,0.0000086174,0.01000,2.695
|
| 45 |
+
6,self_attn.q_proj,0.0000091944,0.01000,2.721
|
| 46 |
+
6,self_attn.v_proj,0.0000042552,0.01000,2.726
|
| 47 |
+
6,self_attn.o_proj,0.0000027771,0.01000,0.525
|
| 48 |
+
6,mlp.gate_proj,0.0000089703,0.01000,1.530
|
| 49 |
+
6,mlp.up_proj,0.0000091738,0.01000,1.535
|
| 50 |
+
6,mlp.down_proj,0.0000123453,0.01000,2.247
|
| 51 |
+
7,self_attn.k_proj,0.0000088630,0.01000,2.920
|
| 52 |
+
7,self_attn.v_proj,0.0000043393,0.01000,2.923
|
| 53 |
+
7,self_attn.q_proj,0.0000109915,0.01000,2.936
|
| 54 |
+
7,self_attn.o_proj,0.0000045459,0.01000,0.515
|
| 55 |
+
7,mlp.gate_proj,0.0000103578,0.01000,1.586
|
| 56 |
+
7,mlp.up_proj,0.0000110500,0.01000,1.595
|
| 57 |
+
7,mlp.down_proj,0.0000322558,0.01000,2.241
|
| 58 |
+
8,self_attn.k_proj,0.0000083509,0.01000,2.739
|
| 59 |
+
8,self_attn.q_proj,0.0000084470,0.01000,2.755
|
| 60 |
+
8,self_attn.v_proj,0.0000058059,0.01000,2.756
|
| 61 |
+
8,self_attn.o_proj,0.0000080038,0.01000,0.527
|
| 62 |
+
8,mlp.gate_proj,0.0000122907,0.01000,1.445
|
| 63 |
+
8,mlp.up_proj,0.0000128683,0.01000,1.450
|
| 64 |
+
8,mlp.down_proj,0.0000226056,0.01000,2.250
|
| 65 |
+
9,self_attn.q_proj,0.0000079025,0.01000,2.782
|
| 66 |
+
9,self_attn.k_proj,0.0000078717,0.01000,2.794
|
| 67 |
+
9,self_attn.v_proj,0.0000050526,0.01000,2.800
|
| 68 |
+
9,self_attn.o_proj,0.0000108522,0.01000,0.521
|
| 69 |
+
9,mlp.up_proj,0.0000132328,0.01000,1.519
|
| 70 |
+
9,mlp.gate_proj,0.0000116754,0.01000,1.532
|
| 71 |
+
9,mlp.down_proj,0.0000234720,0.01000,2.265
|
| 72 |
+
10,self_attn.k_proj,0.0000083320,0.01000,2.716
|
| 73 |
+
10,self_attn.q_proj,0.0000085310,0.01000,2.730
|
| 74 |
+
10,self_attn.v_proj,0.0000051919,0.01000,2.733
|
| 75 |
+
10,self_attn.o_proj,0.0000132269,0.01000,0.517
|
| 76 |
+
10,mlp.up_proj,0.0000129345,0.01000,1.568
|
| 77 |
+
10,mlp.gate_proj,0.0000115324,0.01000,1.570
|
| 78 |
+
10,mlp.down_proj,0.0000272559,0.01000,2.245
|
| 79 |
+
11,self_attn.k_proj,0.0000082962,0.01000,2.772
|
| 80 |
+
11,self_attn.v_proj,0.0000050934,0.01000,2.793
|
| 81 |
+
11,self_attn.q_proj,0.0000085835,0.01000,2.799
|
| 82 |
+
11,self_attn.o_proj,0.0000143053,0.01000,0.511
|
| 83 |
+
11,mlp.gate_proj,0.0000114731,0.01000,1.579
|
| 84 |
+
11,mlp.up_proj,0.0000127581,0.01000,1.582
|
| 85 |
+
11,mlp.down_proj,0.0000286920,0.01000,2.262
|
| 86 |
+
12,self_attn.v_proj,0.0000072761,0.01000,2.874
|
| 87 |
+
12,self_attn.q_proj,0.0000096691,0.01000,2.889
|
| 88 |
+
12,self_attn.k_proj,0.0000090871,0.01000,2.894
|
| 89 |
+
12,self_attn.o_proj,0.0000188947,0.01000,0.525
|
| 90 |
+
12,mlp.up_proj,0.0000133524,0.01000,1.517
|
| 91 |
+
12,mlp.gate_proj,0.0000119858,0.01000,1.528
|
| 92 |
+
12,mlp.down_proj,0.0000316195,0.01000,2.245
|
| 93 |
+
13,self_attn.q_proj,0.0000092859,0.01000,2.695
|
| 94 |
+
13,self_attn.v_proj,0.0000064355,0.01000,2.704
|
| 95 |
+
13,self_attn.k_proj,0.0000089051,0.01000,2.707
|
| 96 |
+
13,self_attn.o_proj,0.0000162895,0.01000,0.523
|
| 97 |
+
13,mlp.gate_proj,0.0000122254,0.01000,1.578
|
| 98 |
+
13,mlp.up_proj,0.0000137294,0.01000,1.586
|
| 99 |
+
13,mlp.down_proj,0.0000361371,0.01000,2.263
|
| 100 |
+
14,self_attn.k_proj,0.0000088439,0.01000,2.775
|
| 101 |
+
14,self_attn.v_proj,0.0000069371,0.01000,2.784
|
| 102 |
+
14,self_attn.q_proj,0.0000087501,0.01000,2.790
|
| 103 |
+
14,self_attn.o_proj,0.0000207656,0.01000,0.517
|
| 104 |
+
14,mlp.up_proj,0.0000135734,0.01000,1.495
|
| 105 |
+
14,mlp.gate_proj,0.0000118168,0.01000,1.497
|
| 106 |
+
14,mlp.down_proj,0.0000438405,0.01000,2.241
|
| 107 |
+
15,self_attn.k_proj,0.0000093432,0.01000,2.750
|
| 108 |
+
15,self_attn.v_proj,0.0000106228,0.01000,2.750
|
| 109 |
+
15,self_attn.q_proj,0.0000096096,0.01000,2.756
|
| 110 |
+
15,self_attn.o_proj,0.0000241663,0.01000,0.517
|
| 111 |
+
15,mlp.up_proj,0.0000146526,0.01000,1.446
|
| 112 |
+
15,mlp.gate_proj,0.0000125828,0.01000,1.450
|
| 113 |
+
15,mlp.down_proj,0.0000627229,0.01000,2.244
|
| 114 |
+
16,self_attn.k_proj,0.0000094416,0.01000,2.723
|
| 115 |
+
16,self_attn.v_proj,0.0000095092,0.01000,2.724
|
| 116 |
+
16,self_attn.q_proj,0.0000094699,0.01000,2.728
|
| 117 |
+
16,self_attn.o_proj,0.0000290566,0.01000,0.518
|
| 118 |
+
16,mlp.up_proj,0.0000180498,0.01000,1.544
|
| 119 |
+
16,mlp.gate_proj,0.0000152804,0.01000,1.550
|
| 120 |
+
16,mlp.down_proj,0.0001009062,0.01000,2.251
|
| 121 |
+
17,self_attn.q_proj,0.0000093712,0.01000,2.720
|
| 122 |
+
17,self_attn.v_proj,0.0000134105,0.01000,2.718
|
| 123 |
+
17,self_attn.k_proj,0.0000093067,0.01000,2.723
|
| 124 |
+
17,self_attn.o_proj,0.0000477296,0.01000,0.519
|
| 125 |
+
17,mlp.gate_proj,0.0000191556,0.01000,1.606
|
| 126 |
+
17,mlp.up_proj,0.0000224697,0.01000,1.624
|
| 127 |
+
17,mlp.down_proj,0.0001616599,0.01000,2.266
|
| 128 |
+
18,self_attn.q_proj,0.0000100121,0.01000,2.840
|
| 129 |
+
18,self_attn.v_proj,0.0000162879,0.01000,2.842
|
| 130 |
+
18,self_attn.k_proj,0.0000096032,0.01000,2.855
|
| 131 |
+
18,self_attn.o_proj,0.0000582959,0.01000,0.529
|
| 132 |
+
18,mlp.gate_proj,0.0000244829,0.01000,1.486
|
| 133 |
+
18,mlp.up_proj,0.0000297788,0.01000,1.496
|
| 134 |
+
18,mlp.down_proj,0.0002775468,0.01000,2.363
|
| 135 |
+
19,self_attn.v_proj,0.0000234440,0.01000,2.878
|
| 136 |
+
19,self_attn.k_proj,0.0000105632,0.01000,2.903
|
| 137 |
+
19,self_attn.q_proj,0.0000111274,0.01000,2.911
|
| 138 |
+
19,self_attn.o_proj,0.0000693788,0.01000,0.535
|
| 139 |
+
19,mlp.up_proj,0.0000364085,0.01000,1.662
|
| 140 |
+
19,mlp.gate_proj,0.0000290227,0.01000,1.687
|
| 141 |
+
19,mlp.down_proj,0.0004459294,0.01000,2.288
|
| 142 |
+
20,self_attn.q_proj,0.0000113580,0.01000,2.919
|
| 143 |
+
20,self_attn.k_proj,0.0000106489,0.01000,2.922
|
| 144 |
+
20,self_attn.v_proj,0.0000282485,0.01000,2.923
|
| 145 |
+
20,self_attn.o_proj,0.0000750161,0.01000,0.520
|
| 146 |
+
20,mlp.gate_proj,0.0000333373,0.01000,1.781
|
| 147 |
+
20,mlp.up_proj,0.0000424925,0.01000,1.796
|
| 148 |
+
20,mlp.down_proj,0.0006084777,0.01000,2.268
|
| 149 |
+
21,self_attn.v_proj,0.0000369492,0.01000,3.099
|
| 150 |
+
21,self_attn.q_proj,0.0000127740,0.01000,3.123
|
| 151 |
+
21,self_attn.k_proj,0.0000122672,0.01000,3.128
|
| 152 |
+
21,self_attn.o_proj,0.0001196352,0.01000,0.524
|
| 153 |
+
21,mlp.up_proj,0.0000489671,0.01000,1.612
|
| 154 |
+
21,mlp.gate_proj,0.0000376844,0.01000,1.621
|
| 155 |
+
21,mlp.down_proj,0.0007977605,0.01000,2.266
|
| 156 |
+
22,self_attn.k_proj,0.0000124738,0.01000,2.829
|
| 157 |
+
22,self_attn.v_proj,0.0000379189,0.01000,2.849
|
| 158 |
+
22,self_attn.q_proj,0.0000131635,0.01000,2.858
|
| 159 |
+
22,self_attn.o_proj,0.0001895542,0.01000,0.519
|
| 160 |
+
22,mlp.up_proj,0.0000555158,0.01000,1.532
|
| 161 |
+
22,mlp.gate_proj,0.0000431185,0.01000,1.545
|
| 162 |
+
22,mlp.down_proj,0.0010612186,0.01000,2.253
|
| 163 |
+
23,self_attn.v_proj,0.0000482519,0.01000,2.967
|
| 164 |
+
23,self_attn.q_proj,0.0000115353,0.01000,2.980
|
| 165 |
+
23,self_attn.k_proj,0.0000119302,0.01000,2.984
|
| 166 |
+
23,self_attn.o_proj,0.0004916409,0.01000,0.519
|
| 167 |
+
23,mlp.gate_proj,0.0000493529,0.01000,1.534
|
| 168 |
+
23,mlp.up_proj,0.0000595984,0.01000,1.539
|
| 169 |
+
23,mlp.down_proj,0.0022477876,0.01000,2.262
|
smollm2-1p7b__gptq__wikitext2__n512__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_mjvln5mc",
|
| 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__s0/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
smollm2-1p7b__gptq__wikitext2__n512__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 |
+
}
|