Upload folder using huggingface_hub
Browse files- smollm2-1p7b__gptq__wikitext2__n128__s2/BENCH_DONE +1 -0
- smollm2-1p7b__gptq__wikitext2__n128__s2/chat_template.jinja +6 -0
- smollm2-1p7b__gptq__wikitext2__n128__s2/config.json +90 -0
- smollm2-1p7b__gptq__wikitext2__n128__s2/generation_config.json +8 -0
- smollm2-1p7b__gptq__wikitext2__n128__s2/model.safetensors +3 -0
- smollm2-1p7b__gptq__wikitext2__n128__s2/quant_log.csv +169 -0
- smollm2-1p7b__gptq__wikitext2__n128__s2/quantize_config.json +47 -0
- smollm2-1p7b__gptq__wikitext2__n128__s2/tokenizer.json +0 -0
- smollm2-1p7b__gptq__wikitext2__n128__s2/tokenizer_config.json +19 -0
smollm2-1p7b__gptq__wikitext2__n128__s2/BENCH_DONE
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
2026-07-31T00:28:54.222767+00:00
|
smollm2-1p7b__gptq__wikitext2__n128__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__n128__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_n90muhuw",
|
| 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__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__n128__s2/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:824fcff7bcbe3928b806b3971cc6512ada8cd6d65f6729e25870e0a714053f64
|
| 3 |
+
size 1040336952
|
smollm2-1p7b__gptq__wikitext2__n128__s2/quant_log.csv
ADDED
|
@@ -0,0 +1,169 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
layer,module,loss,samples,damp,time
|
| 2 |
+
0,self_attn.v_proj,0.0000001315,0.01000,2.833
|
| 3 |
+
0,self_attn.q_proj,0.0000050536,0.01000,2.851
|
| 4 |
+
0,self_attn.k_proj,0.0000037036,0.01000,2.857
|
| 5 |
+
0,self_attn.o_proj,0.0000001095,0.01000,0.526
|
| 6 |
+
0,mlp.up_proj,0.0000051374,0.01000,1.591
|
| 7 |
+
0,mlp.gate_proj,0.0000054511,0.01000,1.603
|
| 8 |
+
0,mlp.down_proj,0.0000028430,0.01000,2.284
|
| 9 |
+
1,self_attn.q_proj,0.0000139278,0.01000,2.839
|
| 10 |
+
1,self_attn.v_proj,0.0000029857,0.01000,2.843
|
| 11 |
+
1,self_attn.k_proj,0.0000124593,0.01000,2.856
|
| 12 |
+
1,self_attn.o_proj,0.0000012206,0.01000,0.531
|
| 13 |
+
1,mlp.up_proj,0.0000089298,0.01000,1.504
|
| 14 |
+
1,mlp.gate_proj,0.0000103282,0.01000,1.514
|
| 15 |
+
1,mlp.down_proj,0.0000442430,0.01000,2.251
|
| 16 |
+
2,self_attn.q_proj,0.0000228562,0.01000,2.653
|
| 17 |
+
2,self_attn.k_proj,0.0000204657,0.01000,2.668
|
| 18 |
+
2,self_attn.v_proj,0.0000080050,0.01000,2.673
|
| 19 |
+
2,self_attn.o_proj,0.0000031883,0.01000,0.518
|
| 20 |
+
2,mlp.up_proj,0.0000142574,0.01000,1.446
|
| 21 |
+
2,mlp.gate_proj,0.0000160240,0.01000,1.447
|
| 22 |
+
2,mlp.down_proj,0.0000120081,0.01000,2.251
|
| 23 |
+
3,self_attn.k_proj,0.0000283570,0.01000,2.693
|
| 24 |
+
3,self_attn.v_proj,0.0000118696,0.01000,2.692
|
| 25 |
+
3,self_attn.q_proj,0.0000282208,0.01000,2.703
|
| 26 |
+
3,self_attn.o_proj,0.0000040169,0.01000,0.514
|
| 27 |
+
3,mlp.gate_proj,0.0000204101,0.01000,1.511
|
| 28 |
+
3,mlp.up_proj,0.0000194039,0.01000,1.515
|
| 29 |
+
3,mlp.down_proj,0.0000197205,0.01000,2.262
|
| 30 |
+
4,self_attn.k_proj,0.0000308712,0.01000,2.804
|
| 31 |
+
4,self_attn.q_proj,0.0000318944,0.01000,2.817
|
| 32 |
+
4,self_attn.v_proj,0.0000134761,0.01000,2.820
|
| 33 |
+
4,self_attn.o_proj,0.0000068281,0.01000,0.522
|
| 34 |
+
4,mlp.up_proj,0.0000247100,0.01000,1.525
|
| 35 |
+
4,mlp.gate_proj,0.0000255343,0.01000,1.538
|
| 36 |
+
4,mlp.down_proj,0.0000284093,0.01000,2.259
|
| 37 |
+
5,self_attn.q_proj,0.0000310237,0.01000,2.709
|
| 38 |
+
5,self_attn.k_proj,0.0000299238,0.01000,2.717
|
| 39 |
+
5,self_attn.v_proj,0.0000137748,0.01000,2.720
|
| 40 |
+
5,self_attn.o_proj,0.0000088568,0.01000,0.520
|
| 41 |
+
5,mlp.gate_proj,0.0000304172,0.01000,1.499
|
| 42 |
+
5,mlp.up_proj,0.0000305402,0.01000,1.506
|
| 43 |
+
5,mlp.down_proj,0.0000364872,0.01000,2.250
|
| 44 |
+
6,self_attn.k_proj,0.0000344456,0.01000,2.712
|
| 45 |
+
6,self_attn.q_proj,0.0000366696,0.01000,2.717
|
| 46 |
+
6,self_attn.v_proj,0.0000170174,0.01000,2.726
|
| 47 |
+
6,self_attn.o_proj,0.0000110498,0.01000,0.516
|
| 48 |
+
6,mlp.gate_proj,0.0000360194,0.01000,1.477
|
| 49 |
+
6,mlp.up_proj,0.0000368625,0.01000,1.479
|
| 50 |
+
6,mlp.down_proj,0.0000488365,0.01000,2.247
|
| 51 |
+
7,self_attn.v_proj,0.0000174310,0.01000,2.953
|
| 52 |
+
7,self_attn.k_proj,0.0000354595,0.01000,2.955
|
| 53 |
+
7,self_attn.q_proj,0.0000439049,0.01000,2.971
|
| 54 |
+
7,self_attn.o_proj,0.0000176574,0.01000,0.518
|
| 55 |
+
7,mlp.up_proj,0.0000448088,0.01000,1.499
|
| 56 |
+
7,mlp.gate_proj,0.0000420172,0.01000,1.512
|
| 57 |
+
7,mlp.down_proj,0.0001264948,0.01000,2.242
|
| 58 |
+
8,self_attn.k_proj,0.0000335635,0.01000,2.728
|
| 59 |
+
8,self_attn.q_proj,0.0000339157,0.01000,2.743
|
| 60 |
+
8,self_attn.v_proj,0.0000233532,0.01000,2.743
|
| 61 |
+
8,self_attn.o_proj,0.0000312559,0.01000,0.518
|
| 62 |
+
8,mlp.up_proj,0.0000518409,0.01000,1.459
|
| 63 |
+
8,mlp.gate_proj,0.0000495097,0.01000,1.475
|
| 64 |
+
8,mlp.down_proj,0.0000901238,0.01000,2.246
|
| 65 |
+
9,self_attn.k_proj,0.0000314996,0.01000,2.727
|
| 66 |
+
9,self_attn.q_proj,0.0000316438,0.01000,2.731
|
| 67 |
+
9,self_attn.v_proj,0.0000202856,0.01000,2.733
|
| 68 |
+
9,self_attn.o_proj,0.0000426982,0.01000,0.516
|
| 69 |
+
9,mlp.gate_proj,0.0000468041,0.01000,1.512
|
| 70 |
+
9,mlp.up_proj,0.0000531612,0.01000,1.518
|
| 71 |
+
9,mlp.down_proj,0.0000926922,0.01000,2.254
|
| 72 |
+
10,self_attn.q_proj,0.0000340312,0.01000,2.716
|
| 73 |
+
10,self_attn.k_proj,0.0000332898,0.01000,2.728
|
| 74 |
+
10,self_attn.v_proj,0.0000208297,0.01000,2.734
|
| 75 |
+
10,self_attn.o_proj,0.0000524593,0.01000,0.516
|
| 76 |
+
10,mlp.up_proj,0.0000516198,0.01000,1.584
|
| 77 |
+
10,mlp.gate_proj,0.0000460696,0.01000,1.595
|
| 78 |
+
10,mlp.down_proj,0.0001066424,0.01000,2.252
|
| 79 |
+
11,self_attn.k_proj,0.0000331351,0.01000,2.720
|
| 80 |
+
11,self_attn.v_proj,0.0000203736,0.01000,2.735
|
| 81 |
+
11,self_attn.q_proj,0.0000342109,0.01000,2.742
|
| 82 |
+
11,self_attn.o_proj,0.0000564186,0.01000,0.519
|
| 83 |
+
11,mlp.gate_proj,0.0000459092,0.01000,1.558
|
| 84 |
+
11,mlp.up_proj,0.0000510197,0.01000,1.562
|
| 85 |
+
11,mlp.down_proj,0.0001131753,0.01000,2.243
|
| 86 |
+
12,self_attn.k_proj,0.0000362623,0.01000,2.746
|
| 87 |
+
12,self_attn.q_proj,0.0000386113,0.01000,2.762
|
| 88 |
+
12,self_attn.v_proj,0.0000291280,0.01000,2.768
|
| 89 |
+
12,self_attn.o_proj,0.0000756645,0.01000,0.515
|
| 90 |
+
12,mlp.gate_proj,0.0000474748,0.01000,1.559
|
| 91 |
+
12,mlp.up_proj,0.0000529605,0.01000,1.559
|
| 92 |
+
12,mlp.down_proj,0.0001225818,0.01000,2.255
|
| 93 |
+
13,self_attn.q_proj,0.0000369540,0.01000,2.922
|
| 94 |
+
13,self_attn.v_proj,0.0000256396,0.01000,2.942
|
| 95 |
+
13,self_attn.k_proj,0.0000353351,0.01000,2.949
|
| 96 |
+
13,self_attn.o_proj,0.0000635183,0.01000,0.520
|
| 97 |
+
13,mlp.gate_proj,0.0000487488,0.01000,1.604
|
| 98 |
+
13,mlp.up_proj,0.0000548038,0.01000,1.611
|
| 99 |
+
13,mlp.down_proj,0.0001422721,0.01000,2.252
|
| 100 |
+
14,self_attn.q_proj,0.0000348595,0.01000,2.694
|
| 101 |
+
14,self_attn.v_proj,0.0000276701,0.01000,2.702
|
| 102 |
+
14,self_attn.k_proj,0.0000352469,0.01000,2.711
|
| 103 |
+
14,self_attn.o_proj,0.0000814746,0.01000,0.515
|
| 104 |
+
14,mlp.gate_proj,0.0000469035,0.01000,1.572
|
| 105 |
+
14,mlp.up_proj,0.0000539995,0.01000,1.583
|
| 106 |
+
14,mlp.down_proj,0.0001713480,0.01000,2.251
|
| 107 |
+
15,self_attn.q_proj,0.0000382854,0.01000,2.747
|
| 108 |
+
15,self_attn.v_proj,0.0000423728,0.01000,2.765
|
| 109 |
+
15,self_attn.k_proj,0.0000371263,0.01000,2.772
|
| 110 |
+
15,self_attn.o_proj,0.0000939422,0.01000,0.506
|
| 111 |
+
15,mlp.up_proj,0.0000583667,0.01000,1.476
|
| 112 |
+
15,mlp.gate_proj,0.0000500885,0.01000,1.482
|
| 113 |
+
15,mlp.down_proj,0.0002465918,0.01000,2.249
|
| 114 |
+
16,self_attn.v_proj,0.0000378443,0.01000,2.687
|
| 115 |
+
16,self_attn.q_proj,0.0000376949,0.01000,2.700
|
| 116 |
+
16,self_attn.k_proj,0.0000375540,0.01000,2.707
|
| 117 |
+
16,self_attn.o_proj,0.0001148094,0.01000,0.512
|
| 118 |
+
16,mlp.gate_proj,0.0000609288,0.01000,1.449
|
| 119 |
+
16,mlp.up_proj,0.0000720348,0.01000,1.460
|
| 120 |
+
16,mlp.down_proj,0.0003987056,0.01000,2.254
|
| 121 |
+
17,self_attn.q_proj,0.0000373106,0.01000,2.609
|
| 122 |
+
17,self_attn.k_proj,0.0000369693,0.01000,2.609
|
| 123 |
+
17,self_attn.v_proj,0.0000533228,0.01000,2.609
|
| 124 |
+
17,self_attn.o_proj,0.0001962984,0.01000,0.517
|
| 125 |
+
17,mlp.gate_proj,0.0000762355,0.01000,1.642
|
| 126 |
+
17,mlp.up_proj,0.0000894323,0.01000,1.646
|
| 127 |
+
17,mlp.down_proj,0.0006327870,0.01000,2.247
|
| 128 |
+
18,self_attn.q_proj,0.0000399231,0.01000,2.782
|
| 129 |
+
18,self_attn.k_proj,0.0000382039,0.01000,2.782
|
| 130 |
+
18,self_attn.v_proj,0.0000647972,0.01000,2.786
|
| 131 |
+
18,self_attn.o_proj,0.0002304173,0.01000,0.521
|
| 132 |
+
18,mlp.gate_proj,0.0000976185,0.01000,1.691
|
| 133 |
+
18,mlp.up_proj,0.0001188175,0.01000,1.699
|
| 134 |
+
18,mlp.down_proj,0.0010829040,0.01000,2.239
|
| 135 |
+
19,self_attn.q_proj,0.0000443792,0.01000,2.690
|
| 136 |
+
19,self_attn.k_proj,0.0000420157,0.01000,2.689
|
| 137 |
+
19,self_attn.v_proj,0.0000931744,0.01000,2.698
|
| 138 |
+
19,self_attn.o_proj,0.0002711860,0.01000,0.513
|
| 139 |
+
19,mlp.up_proj,0.0001450418,0.01000,1.556
|
| 140 |
+
19,mlp.gate_proj,0.0001153745,0.01000,1.566
|
| 141 |
+
19,mlp.down_proj,0.0017201067,0.01000,2.246
|
| 142 |
+
20,self_attn.q_proj,0.0000450008,0.01000,2.852
|
| 143 |
+
20,self_attn.v_proj,0.0001117322,0.01000,2.871
|
| 144 |
+
20,self_attn.k_proj,0.0000420948,0.01000,2.876
|
| 145 |
+
20,self_attn.o_proj,0.0002943531,0.01000,0.511
|
| 146 |
+
20,mlp.up_proj,0.0001687475,0.01000,1.541
|
| 147 |
+
20,mlp.gate_proj,0.0001322693,0.01000,1.552
|
| 148 |
+
20,mlp.down_proj,0.0023366389,0.01000,2.240
|
| 149 |
+
21,self_attn.k_proj,0.0000486096,0.01000,2.809
|
| 150 |
+
21,self_attn.q_proj,0.0000505927,0.01000,2.814
|
| 151 |
+
21,self_attn.v_proj,0.0001461148,0.01000,2.814
|
| 152 |
+
21,self_attn.o_proj,0.0004507883,0.01000,0.631
|
| 153 |
+
21,mlp.up_proj,0.0001940392,0.01000,1.583
|
| 154 |
+
21,mlp.gate_proj,0.0001493154,0.01000,1.596
|
| 155 |
+
21,mlp.down_proj,0.0030648210,0.01000,2.246
|
| 156 |
+
22,self_attn.k_proj,0.0000495095,0.01000,2.628
|
| 157 |
+
22,self_attn.q_proj,0.0000522156,0.01000,2.649
|
| 158 |
+
22,self_attn.v_proj,0.0001505830,0.01000,2.650
|
| 159 |
+
22,self_attn.o_proj,0.0007417422,0.01000,0.515
|
| 160 |
+
22,mlp.up_proj,0.0002206181,0.01000,1.459
|
| 161 |
+
22,mlp.gate_proj,0.0001712538,0.01000,1.464
|
| 162 |
+
22,mlp.down_proj,0.0041234960,0.01000,2.251
|
| 163 |
+
23,self_attn.k_proj,0.0000474103,0.01000,2.700
|
| 164 |
+
23,self_attn.q_proj,0.0000458310,0.01000,2.711
|
| 165 |
+
23,self_attn.v_proj,0.0001920256,0.01000,2.714
|
| 166 |
+
23,self_attn.o_proj,0.0019186277,0.01000,0.510
|
| 167 |
+
23,mlp.gate_proj,0.0001959021,0.01000,1.503
|
| 168 |
+
23,mlp.up_proj,0.0002367566,0.01000,1.514
|
| 169 |
+
23,mlp.down_proj,0.0087288471,0.01000,2.248
|
smollm2-1p7b__gptq__wikitext2__n128__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_n90muhuw",
|
| 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__s2/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
smollm2-1p7b__gptq__wikitext2__n128__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 |
+
}
|