Upload folder using huggingface_hub
Browse files- smollm2-1p7b__gptq__wikitext2__n128__s1/BENCH_DONE +1 -0
- smollm2-1p7b__gptq__wikitext2__n128__s1/chat_template.jinja +6 -0
- smollm2-1p7b__gptq__wikitext2__n128__s1/config.json +90 -0
- smollm2-1p7b__gptq__wikitext2__n128__s1/generation_config.json +8 -0
- smollm2-1p7b__gptq__wikitext2__n128__s1/model.safetensors +3 -0
- smollm2-1p7b__gptq__wikitext2__n128__s1/quant_log.csv +169 -0
- smollm2-1p7b__gptq__wikitext2__n128__s1/quantize_config.json +47 -0
- smollm2-1p7b__gptq__wikitext2__n128__s1/tokenizer.json +0 -0
- smollm2-1p7b__gptq__wikitext2__n128__s1/tokenizer_config.json +19 -0
smollm2-1p7b__gptq__wikitext2__n128__s1/BENCH_DONE
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
2026-07-31T00:22:10.326104+00:00
|
smollm2-1p7b__gptq__wikitext2__n128__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__n128__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_82gz4k5s",
|
| 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__n128__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__n128__s1/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:04ad8377807bf1eca1409598348b908f9c200b21d5ee6c65a954968bc884ef84
|
| 3 |
+
size 1040336952
|
smollm2-1p7b__gptq__wikitext2__n128__s1/quant_log.csv
ADDED
|
@@ -0,0 +1,169 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
layer,module,loss,samples,damp,time
|
| 2 |
+
0,self_attn.v_proj,0.0000001287,0.01000,3.133
|
| 3 |
+
0,self_attn.k_proj,0.0000036351,0.01000,3.148
|
| 4 |
+
0,self_attn.q_proj,0.0000049473,0.01000,3.159
|
| 5 |
+
0,self_attn.o_proj,0.0000001102,0.01000,0.525
|
| 6 |
+
0,mlp.up_proj,0.0000050381,0.01000,1.640
|
| 7 |
+
0,mlp.gate_proj,0.0000053506,0.01000,1.650
|
| 8 |
+
0,mlp.down_proj,0.0000027895,0.01000,2.259
|
| 9 |
+
1,self_attn.q_proj,0.0000138396,0.01000,2.923
|
| 10 |
+
1,self_attn.k_proj,0.0000123499,0.01000,2.933
|
| 11 |
+
1,self_attn.v_proj,0.0000029527,0.01000,2.936
|
| 12 |
+
1,self_attn.o_proj,0.0000012230,0.01000,0.525
|
| 13 |
+
1,mlp.up_proj,0.0000088470,0.01000,1.599
|
| 14 |
+
1,mlp.gate_proj,0.0000102190,0.01000,1.608
|
| 15 |
+
1,mlp.down_proj,0.0000454952,0.01000,2.262
|
| 16 |
+
2,self_attn.q_proj,0.0000226839,0.01000,3.114
|
| 17 |
+
2,self_attn.k_proj,0.0000202795,0.01000,3.120
|
| 18 |
+
2,self_attn.v_proj,0.0000079351,0.01000,3.126
|
| 19 |
+
2,self_attn.o_proj,0.0000031360,0.01000,0.513
|
| 20 |
+
2,mlp.up_proj,0.0000140397,0.01000,1.443
|
| 21 |
+
2,mlp.gate_proj,0.0000157769,0.01000,1.449
|
| 22 |
+
2,mlp.down_proj,0.0000115634,0.01000,2.247
|
| 23 |
+
3,self_attn.q_proj,0.0000279007,0.01000,2.866
|
| 24 |
+
3,self_attn.v_proj,0.0000116954,0.01000,2.868
|
| 25 |
+
3,self_attn.k_proj,0.0000279987,0.01000,2.867
|
| 26 |
+
3,self_attn.o_proj,0.0000039384,0.01000,0.513
|
| 27 |
+
3,mlp.gate_proj,0.0000200172,0.01000,1.481
|
| 28 |
+
3,mlp.up_proj,0.0000190241,0.01000,1.485
|
| 29 |
+
3,mlp.down_proj,0.0000189716,0.01000,2.240
|
| 30 |
+
4,self_attn.k_proj,0.0000303030,0.01000,2.821
|
| 31 |
+
4,self_attn.q_proj,0.0000313628,0.01000,2.827
|
| 32 |
+
4,self_attn.v_proj,0.0000132143,0.01000,2.832
|
| 33 |
+
4,self_attn.o_proj,0.0000067843,0.01000,0.521
|
| 34 |
+
4,mlp.gate_proj,0.0000248752,0.01000,1.564
|
| 35 |
+
4,mlp.up_proj,0.0000240593,0.01000,1.576
|
| 36 |
+
4,mlp.down_proj,0.0000269302,0.01000,2.233
|
| 37 |
+
5,self_attn.q_proj,0.0000303804,0.01000,3.091
|
| 38 |
+
5,self_attn.v_proj,0.0000134855,0.01000,3.114
|
| 39 |
+
5,self_attn.k_proj,0.0000293331,0.01000,3.119
|
| 40 |
+
5,self_attn.o_proj,0.0000087682,0.01000,0.520
|
| 41 |
+
5,mlp.gate_proj,0.0000296132,0.01000,1.488
|
| 42 |
+
5,mlp.up_proj,0.0000298002,0.01000,1.496
|
| 43 |
+
5,mlp.down_proj,0.0000350293,0.01000,2.246
|
| 44 |
+
6,self_attn.v_proj,0.0000167096,0.01000,3.055
|
| 45 |
+
6,self_attn.k_proj,0.0000338393,0.01000,3.069
|
| 46 |
+
6,self_attn.q_proj,0.0000360446,0.01000,3.076
|
| 47 |
+
6,self_attn.o_proj,0.0000110559,0.01000,0.520
|
| 48 |
+
6,mlp.up_proj,0.0000359598,0.01000,1.477
|
| 49 |
+
6,mlp.gate_proj,0.0000351025,0.01000,1.482
|
| 50 |
+
6,mlp.down_proj,0.0000469828,0.01000,2.241
|
| 51 |
+
7,self_attn.v_proj,0.0000171332,0.01000,2.889
|
| 52 |
+
7,self_attn.k_proj,0.0000349175,0.01000,2.896
|
| 53 |
+
7,self_attn.q_proj,0.0000433119,0.01000,2.908
|
| 54 |
+
7,self_attn.o_proj,0.0000176624,0.01000,0.513
|
| 55 |
+
7,mlp.gate_proj,0.0000409533,0.01000,1.802
|
| 56 |
+
7,mlp.up_proj,0.0000437307,0.01000,1.815
|
| 57 |
+
7,mlp.down_proj,0.0001177320,0.01000,2.252
|
| 58 |
+
8,self_attn.v_proj,0.0000229655,0.01000,2.950
|
| 59 |
+
8,self_attn.k_proj,0.0000330289,0.01000,2.961
|
| 60 |
+
8,self_attn.q_proj,0.0000333941,0.01000,2.965
|
| 61 |
+
8,self_attn.o_proj,0.0000306277,0.01000,0.516
|
| 62 |
+
8,mlp.gate_proj,0.0000485873,0.01000,1.752
|
| 63 |
+
8,mlp.up_proj,0.0000508907,0.01000,1.763
|
| 64 |
+
8,mlp.down_proj,0.0000870766,0.01000,2.244
|
| 65 |
+
9,self_attn.k_proj,0.0000311102,0.01000,2.788
|
| 66 |
+
9,self_attn.v_proj,0.0000199893,0.01000,2.792
|
| 67 |
+
9,self_attn.q_proj,0.0000312271,0.01000,2.807
|
| 68 |
+
9,self_attn.o_proj,0.0000418638,0.01000,0.526
|
| 69 |
+
9,mlp.up_proj,0.0000522324,0.01000,1.603
|
| 70 |
+
9,mlp.gate_proj,0.0000460634,0.01000,1.610
|
| 71 |
+
9,mlp.down_proj,0.0000900177,0.01000,2.246
|
| 72 |
+
10,self_attn.k_proj,0.0000330314,0.01000,2.658
|
| 73 |
+
10,self_attn.q_proj,0.0000336916,0.01000,2.670
|
| 74 |
+
10,self_attn.v_proj,0.0000205761,0.01000,2.674
|
| 75 |
+
10,self_attn.o_proj,0.0000528915,0.01000,0.511
|
| 76 |
+
10,mlp.gate_proj,0.0000452965,0.01000,1.449
|
| 77 |
+
10,mlp.up_proj,0.0000507293,0.01000,1.458
|
| 78 |
+
10,mlp.down_proj,0.0001041150,0.01000,2.244
|
| 79 |
+
11,self_attn.q_proj,0.0000339143,0.01000,2.593
|
| 80 |
+
11,self_attn.k_proj,0.0000327819,0.01000,2.608
|
| 81 |
+
11,self_attn.v_proj,0.0000201354,0.01000,2.611
|
| 82 |
+
11,self_attn.o_proj,0.0000570022,0.01000,0.510
|
| 83 |
+
11,mlp.gate_proj,0.0000448873,0.01000,1.497
|
| 84 |
+
11,mlp.up_proj,0.0000498736,0.01000,1.505
|
| 85 |
+
11,mlp.down_proj,0.0001093105,0.01000,2.235
|
| 86 |
+
12,self_attn.v_proj,0.0000288429,0.01000,2.777
|
| 87 |
+
12,self_attn.q_proj,0.0000382820,0.01000,2.792
|
| 88 |
+
12,self_attn.k_proj,0.0000359018,0.01000,2.796
|
| 89 |
+
12,self_attn.o_proj,0.0000775334,0.01000,0.518
|
| 90 |
+
12,mlp.up_proj,0.0000522739,0.01000,1.495
|
| 91 |
+
12,mlp.gate_proj,0.0000468500,0.01000,1.503
|
| 92 |
+
12,mlp.down_proj,0.0001210542,0.01000,2.257
|
| 93 |
+
13,self_attn.q_proj,0.0000366947,0.01000,2.748
|
| 94 |
+
13,self_attn.k_proj,0.0000351594,0.01000,2.761
|
| 95 |
+
13,self_attn.v_proj,0.0000254146,0.01000,2.763
|
| 96 |
+
13,self_attn.o_proj,0.0000641707,0.01000,0.518
|
| 97 |
+
13,mlp.gate_proj,0.0000479099,0.01000,1.566
|
| 98 |
+
13,mlp.up_proj,0.0000538420,0.01000,1.565
|
| 99 |
+
13,mlp.down_proj,0.0001396372,0.01000,2.245
|
| 100 |
+
14,self_attn.k_proj,0.0000351832,0.01000,2.807
|
| 101 |
+
14,self_attn.q_proj,0.0000347396,0.01000,2.819
|
| 102 |
+
14,self_attn.v_proj,0.0000274799,0.01000,2.821
|
| 103 |
+
14,self_attn.o_proj,0.0000815302,0.01000,0.518
|
| 104 |
+
14,mlp.gate_proj,0.0000462022,0.01000,1.499
|
| 105 |
+
14,mlp.up_proj,0.0000530818,0.01000,1.501
|
| 106 |
+
14,mlp.down_proj,0.0001676278,0.01000,2.251
|
| 107 |
+
15,self_attn.k_proj,0.0000372603,0.01000,2.787
|
| 108 |
+
15,self_attn.q_proj,0.0000383167,0.01000,2.800
|
| 109 |
+
15,self_attn.v_proj,0.0000424176,0.01000,2.801
|
| 110 |
+
15,self_attn.o_proj,0.0000957277,0.01000,0.514
|
| 111 |
+
15,mlp.gate_proj,0.0000498395,0.01000,1.561
|
| 112 |
+
15,mlp.up_proj,0.0000581153,0.01000,1.568
|
| 113 |
+
15,mlp.down_proj,0.0002431892,0.01000,2.249
|
| 114 |
+
16,self_attn.v_proj,0.0000379547,0.01000,2.653
|
| 115 |
+
16,self_attn.k_proj,0.0000376430,0.01000,2.656
|
| 116 |
+
16,self_attn.q_proj,0.0000378731,0.01000,2.666
|
| 117 |
+
16,self_attn.o_proj,0.0001156948,0.01000,0.522
|
| 118 |
+
16,mlp.gate_proj,0.0000605396,0.01000,1.422
|
| 119 |
+
16,mlp.up_proj,0.0000715840,0.01000,1.430
|
| 120 |
+
16,mlp.down_proj,0.0003943855,0.01000,2.229
|
| 121 |
+
17,self_attn.v_proj,0.0000532933,0.01000,2.644
|
| 122 |
+
17,self_attn.k_proj,0.0000369317,0.01000,2.658
|
| 123 |
+
17,self_attn.q_proj,0.0000371970,0.01000,2.660
|
| 124 |
+
17,self_attn.o_proj,0.0001922159,0.01000,0.518
|
| 125 |
+
17,mlp.up_proj,0.0000884403,0.01000,1.477
|
| 126 |
+
17,mlp.gate_proj,0.0000753017,0.01000,1.480
|
| 127 |
+
17,mlp.down_proj,0.0006146104,0.01000,2.229
|
| 128 |
+
18,self_attn.k_proj,0.0000379321,0.01000,2.630
|
| 129 |
+
18,self_attn.q_proj,0.0000396400,0.01000,2.632
|
| 130 |
+
18,self_attn.v_proj,0.0000644059,0.01000,2.643
|
| 131 |
+
18,self_attn.o_proj,0.0002345307,0.01000,0.511
|
| 132 |
+
18,mlp.up_proj,0.0001173110,0.01000,1.499
|
| 133 |
+
18,mlp.gate_proj,0.0000962752,0.01000,1.507
|
| 134 |
+
18,mlp.down_proj,0.0010544495,0.01000,2.228
|
| 135 |
+
19,self_attn.q_proj,0.0000437249,0.01000,2.715
|
| 136 |
+
19,self_attn.v_proj,0.0000919667,0.01000,2.726
|
| 137 |
+
19,self_attn.k_proj,0.0000414613,0.01000,2.733
|
| 138 |
+
19,self_attn.o_proj,0.0002694627,0.01000,0.512
|
| 139 |
+
19,mlp.up_proj,0.0001421104,0.01000,1.539
|
| 140 |
+
19,mlp.gate_proj,0.0001130587,0.01000,1.543
|
| 141 |
+
19,mlp.down_proj,0.0016527406,0.01000,2.238
|
| 142 |
+
20,self_attn.k_proj,0.0000413833,0.01000,2.821
|
| 143 |
+
20,self_attn.v_proj,0.0001096809,0.01000,2.835
|
| 144 |
+
20,self_attn.q_proj,0.0000441594,0.01000,2.844
|
| 145 |
+
20,self_attn.o_proj,0.0002836728,0.01000,0.514
|
| 146 |
+
20,mlp.gate_proj,0.0001290659,0.01000,1.601
|
| 147 |
+
20,mlp.up_proj,0.0001646571,0.01000,1.609
|
| 148 |
+
20,mlp.down_proj,0.0022439079,0.01000,2.246
|
| 149 |
+
21,self_attn.q_proj,0.0000495705,0.01000,2.713
|
| 150 |
+
21,self_attn.v_proj,0.0001433593,0.01000,2.729
|
| 151 |
+
21,self_attn.k_proj,0.0000476724,0.01000,2.732
|
| 152 |
+
21,self_attn.o_proj,0.0004578074,0.01000,0.514
|
| 153 |
+
21,mlp.gate_proj,0.0001455639,0.01000,1.385
|
| 154 |
+
21,mlp.up_proj,0.0001890752,0.01000,1.386
|
| 155 |
+
21,mlp.down_proj,0.0029423840,0.01000,2.229
|
| 156 |
+
22,self_attn.v_proj,0.0001481852,0.01000,2.836
|
| 157 |
+
22,self_attn.k_proj,0.0000487704,0.01000,2.857
|
| 158 |
+
22,self_attn.q_proj,0.0000513559,0.01000,2.860
|
| 159 |
+
22,self_attn.o_proj,0.0007317386,0.01000,0.512
|
| 160 |
+
22,mlp.up_proj,0.0002143792,0.01000,1.513
|
| 161 |
+
22,mlp.gate_proj,0.0001664193,0.01000,1.528
|
| 162 |
+
22,mlp.down_proj,0.0038099554,0.01000,2.240
|
| 163 |
+
23,self_attn.k_proj,0.0000464830,0.01000,2.757
|
| 164 |
+
23,self_attn.v_proj,0.0001884275,0.01000,2.770
|
| 165 |
+
23,self_attn.q_proj,0.0000449397,0.01000,2.779
|
| 166 |
+
23,self_attn.o_proj,0.0018054827,0.01000,0.519
|
| 167 |
+
23,mlp.up_proj,0.0002294396,0.01000,1.551
|
| 168 |
+
23,mlp.gate_proj,0.0001899644,0.01000,1.555
|
| 169 |
+
23,mlp.down_proj,0.0084354896,0.01000,2.239
|
smollm2-1p7b__gptq__wikitext2__n128__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_82gz4k5s",
|
| 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__n128__s1/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
smollm2-1p7b__gptq__wikitext2__n128__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 |
+
}
|