Mohaaxa commited on
Commit
74ad13f
·
verified ·
1 Parent(s): fd6ee5c

Upload folder using huggingface_hub

Browse files
smollm2-1p7b__gptq__wikitext2__n512__s1/BENCH_DONE ADDED
@@ -0,0 +1 @@
 
 
1
+ 2026-07-31T00:57:21.681695+00:00
smollm2-1p7b__gptq__wikitext2__n512__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__n512__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_49i1jpy7",
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__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__n512__s1/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c78c8409139030c3e155a97692d5437aa988163d819e0bacc29e2267c76b0b55
3
+ size 1040336952
smollm2-1p7b__gptq__wikitext2__n512__s1/quant_log.csv ADDED
@@ -0,0 +1,169 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ layer,module,loss,samples,damp,time
2
+ 0,self_attn.k_proj,0.0000009407,0.01000,2.886
3
+ 0,self_attn.v_proj,0.0000000332,0.01000,2.892
4
+ 0,self_attn.q_proj,0.0000012832,0.01000,2.902
5
+ 0,self_attn.o_proj,0.0000000276,0.01000,0.517
6
+ 0,mlp.up_proj,0.0000012989,0.01000,1.670
7
+ 0,mlp.gate_proj,0.0000013782,0.01000,1.674
8
+ 0,mlp.down_proj,0.0000007689,0.01000,2.229
9
+ 1,self_attn.q_proj,0.0000035067,0.01000,2.867
10
+ 1,self_attn.k_proj,0.0000031335,0.01000,2.871
11
+ 1,self_attn.v_proj,0.0000007500,0.01000,2.878
12
+ 1,self_attn.o_proj,0.0000003183,0.01000,0.510
13
+ 1,mlp.up_proj,0.0000022446,0.01000,1.527
14
+ 1,mlp.gate_proj,0.0000025958,0.01000,1.536
15
+ 1,mlp.down_proj,0.0000113333,0.01000,2.293
16
+ 2,self_attn.q_proj,0.0000057572,0.01000,2.682
17
+ 2,self_attn.k_proj,0.0000051432,0.01000,2.682
18
+ 2,self_attn.v_proj,0.0000020116,0.01000,2.690
19
+ 2,self_attn.o_proj,0.0000008030,0.01000,0.527
20
+ 2,mlp.gate_proj,0.0000040385,0.01000,1.509
21
+ 2,mlp.up_proj,0.0000035947,0.01000,1.508
22
+ 2,mlp.down_proj,0.0000031073,0.01000,2.237
23
+ 3,self_attn.v_proj,0.0000029847,0.01000,3.005
24
+ 3,self_attn.k_proj,0.0000071348,0.01000,3.015
25
+ 3,self_attn.q_proj,0.0000071198,0.01000,3.024
26
+ 3,self_attn.o_proj,0.0000010339,0.01000,0.517
27
+ 3,mlp.up_proj,0.0000048457,0.01000,1.568
28
+ 3,mlp.gate_proj,0.0000050972,0.01000,1.570
29
+ 3,mlp.down_proj,0.0000049984,0.01000,2.246
30
+ 4,self_attn.v_proj,0.0000033596,0.01000,3.121
31
+ 4,self_attn.k_proj,0.0000077025,0.01000,3.144
32
+ 4,self_attn.q_proj,0.0000079704,0.01000,3.156
33
+ 4,self_attn.o_proj,0.0000017346,0.01000,0.523
34
+ 4,mlp.up_proj,0.0000061561,0.01000,1.593
35
+ 4,mlp.gate_proj,0.0000063587,0.01000,1.600
36
+ 4,mlp.down_proj,0.0000071457,0.01000,2.259
37
+ 5,self_attn.v_proj,0.0000034258,0.01000,2.837
38
+ 5,self_attn.k_proj,0.0000074511,0.01000,2.858
39
+ 5,self_attn.q_proj,0.0000077249,0.01000,2.865
40
+ 5,self_attn.o_proj,0.0000021998,0.01000,0.520
41
+ 5,mlp.up_proj,0.0000075891,0.01000,1.562
42
+ 5,mlp.gate_proj,0.0000075475,0.01000,1.574
43
+ 5,mlp.down_proj,0.0000092213,0.01000,2.253
44
+ 6,self_attn.q_proj,0.0000091388,0.01000,2.803
45
+ 6,self_attn.v_proj,0.0000042310,0.01000,2.814
46
+ 6,self_attn.k_proj,0.0000085659,0.01000,2.821
47
+ 6,self_attn.o_proj,0.0000027761,0.01000,0.514
48
+ 6,mlp.gate_proj,0.0000089347,0.01000,1.581
49
+ 6,mlp.up_proj,0.0000091434,0.01000,1.588
50
+ 6,mlp.down_proj,0.0000123697,0.01000,2.249
51
+ 7,self_attn.v_proj,0.0000043304,0.01000,2.769
52
+ 7,self_attn.k_proj,0.0000088238,0.01000,2.776
53
+ 7,self_attn.q_proj,0.0000109664,0.01000,2.794
54
+ 7,self_attn.o_proj,0.0000044583,0.01000,0.538
55
+ 7,mlp.gate_proj,0.0000104132,0.01000,1.506
56
+ 7,mlp.up_proj,0.0000111148,0.01000,1.521
57
+ 7,mlp.down_proj,0.0000328380,0.01000,2.257
58
+ 8,self_attn.v_proj,0.0000058130,0.01000,2.956
59
+ 8,self_attn.q_proj,0.0000084475,0.01000,2.973
60
+ 8,self_attn.k_proj,0.0000083589,0.01000,2.977
61
+ 8,self_attn.o_proj,0.0000081063,0.01000,0.518
62
+ 8,mlp.up_proj,0.0000128970,0.01000,1.575
63
+ 8,mlp.gate_proj,0.0000123231,0.01000,1.585
64
+ 8,mlp.down_proj,0.0000230025,0.01000,2.239
65
+ 9,self_attn.q_proj,0.0000079297,0.01000,2.818
66
+ 9,self_attn.v_proj,0.0000050742,0.01000,2.831
67
+ 9,self_attn.k_proj,0.0000078928,0.01000,2.836
68
+ 9,self_attn.o_proj,0.0000109994,0.01000,0.518
69
+ 9,mlp.gate_proj,0.0000117093,0.01000,1.532
70
+ 9,mlp.up_proj,0.0000132917,0.01000,1.542
71
+ 9,mlp.down_proj,0.0000236418,0.01000,2.305
72
+ 10,self_attn.q_proj,0.0000085552,0.01000,2.789
73
+ 10,self_attn.v_proj,0.0000052120,0.01000,2.787
74
+ 10,self_attn.k_proj,0.0000083618,0.01000,2.802
75
+ 10,self_attn.o_proj,0.0000133981,0.01000,0.519
76
+ 10,mlp.gate_proj,0.0000115535,0.01000,1.475
77
+ 10,mlp.up_proj,0.0000129360,0.01000,1.481
78
+ 10,mlp.down_proj,0.0000272613,0.01000,2.246
79
+ 11,self_attn.k_proj,0.0000083127,0.01000,2.901
80
+ 11,self_attn.v_proj,0.0000051083,0.01000,2.912
81
+ 11,self_attn.q_proj,0.0000085827,0.01000,2.925
82
+ 11,self_attn.o_proj,0.0000144886,0.01000,0.547
83
+ 11,mlp.up_proj,0.0000127611,0.01000,1.675
84
+ 11,mlp.gate_proj,0.0000114931,0.01000,1.685
85
+ 11,mlp.down_proj,0.0000285565,0.01000,2.284
86
+ 12,self_attn.v_proj,0.0000072877,0.01000,2.881
87
+ 12,self_attn.q_proj,0.0000096831,0.01000,2.902
88
+ 12,self_attn.k_proj,0.0000091093,0.01000,2.907
89
+ 12,self_attn.o_proj,0.0000194640,0.01000,0.519
90
+ 12,mlp.gate_proj,0.0000119416,0.01000,1.568
91
+ 12,mlp.up_proj,0.0000133065,0.01000,1.571
92
+ 12,mlp.down_proj,0.0000314858,0.01000,2.249
93
+ 13,self_attn.q_proj,0.0000093066,0.01000,3.025
94
+ 13,self_attn.v_proj,0.0000064457,0.01000,3.044
95
+ 13,self_attn.k_proj,0.0000089090,0.01000,3.046
96
+ 13,self_attn.o_proj,0.0000164238,0.01000,0.518
97
+ 13,mlp.gate_proj,0.0000122293,0.01000,1.506
98
+ 13,mlp.up_proj,0.0000137359,0.01000,1.515
99
+ 13,mlp.down_proj,0.0000361997,0.01000,2.267
100
+ 14,self_attn.q_proj,0.0000087863,0.01000,3.026
101
+ 14,self_attn.k_proj,0.0000088744,0.01000,3.036
102
+ 14,self_attn.v_proj,0.0000069522,0.01000,3.040
103
+ 14,self_attn.o_proj,0.0000208707,0.01000,0.518
104
+ 14,mlp.up_proj,0.0000135752,0.01000,1.546
105
+ 14,mlp.gate_proj,0.0000118138,0.01000,1.556
106
+ 14,mlp.down_proj,0.0000438524,0.01000,2.239
107
+ 15,self_attn.k_proj,0.0000093908,0.01000,2.870
108
+ 15,self_attn.q_proj,0.0000096660,0.01000,2.919
109
+ 15,self_attn.v_proj,0.0000106840,0.01000,2.920
110
+ 15,self_attn.o_proj,0.0000244707,0.01000,0.519
111
+ 15,mlp.gate_proj,0.0000126271,0.01000,1.690
112
+ 15,mlp.up_proj,0.0000147105,0.01000,1.703
113
+ 15,mlp.down_proj,0.0000628941,0.01000,2.276
114
+ 16,self_attn.k_proj,0.0000094881,0.01000,2.762
115
+ 16,self_attn.v_proj,0.0000095635,0.01000,2.766
116
+ 16,self_attn.q_proj,0.0000095247,0.01000,2.779
117
+ 16,self_attn.o_proj,0.0000288576,0.01000,0.521
118
+ 16,mlp.gate_proj,0.0000154086,0.01000,1.584
119
+ 16,mlp.up_proj,0.0000182085,0.01000,1.597
120
+ 16,mlp.down_proj,0.0001016312,0.01000,2.251
121
+ 17,self_attn.v_proj,0.0000134437,0.01000,2.885
122
+ 17,self_attn.q_proj,0.0000094027,0.01000,2.899
123
+ 17,self_attn.k_proj,0.0000093239,0.01000,2.909
124
+ 17,self_attn.o_proj,0.0000487763,0.01000,0.521
125
+ 17,mlp.gate_proj,0.0000191970,0.01000,1.591
126
+ 17,mlp.up_proj,0.0000225227,0.01000,1.597
127
+ 17,mlp.down_proj,0.0001612001,0.01000,2.241
128
+ 18,self_attn.q_proj,0.0000100815,0.01000,2.811
129
+ 18,self_attn.v_proj,0.0000163536,0.01000,2.824
130
+ 18,self_attn.k_proj,0.0000096453,0.01000,2.829
131
+ 18,self_attn.o_proj,0.0000600355,0.01000,0.507
132
+ 18,mlp.up_proj,0.0000298414,0.01000,1.543
133
+ 18,mlp.gate_proj,0.0000245265,0.01000,1.553
134
+ 18,mlp.down_proj,0.0002758192,0.01000,2.244
135
+ 19,self_attn.k_proj,0.0000106073,0.01000,2.832
136
+ 19,self_attn.v_proj,0.0000235388,0.01000,2.833
137
+ 19,self_attn.q_proj,0.0000111896,0.01000,2.844
138
+ 19,self_attn.o_proj,0.0000705977,0.01000,0.531
139
+ 19,mlp.up_proj,0.0000365080,0.01000,1.575
140
+ 19,mlp.gate_proj,0.0000290669,0.01000,1.583
141
+ 19,mlp.down_proj,0.0004444962,0.01000,2.241
142
+ 20,self_attn.v_proj,0.0000283949,0.01000,2.903
143
+ 20,self_attn.k_proj,0.0000107144,0.01000,2.918
144
+ 20,self_attn.q_proj,0.0000114247,0.01000,2.923
145
+ 20,self_attn.o_proj,0.0000758717,0.01000,0.583
146
+ 20,mlp.up_proj,0.0000425666,0.01000,1.603
147
+ 20,mlp.gate_proj,0.0000333607,0.01000,1.608
148
+ 20,mlp.down_proj,0.0006041649,0.01000,2.253
149
+ 21,self_attn.q_proj,0.0000128293,0.01000,2.755
150
+ 21,self_attn.v_proj,0.0000370449,0.01000,2.766
151
+ 21,self_attn.k_proj,0.0000123291,0.01000,2.771
152
+ 21,self_attn.o_proj,0.0001194729,0.01000,0.513
153
+ 21,mlp.gate_proj,0.0000376966,0.01000,1.525
154
+ 21,mlp.up_proj,0.0000489228,0.01000,1.539
155
+ 21,mlp.down_proj,0.0007893776,0.01000,2.248
156
+ 22,self_attn.v_proj,0.0000381653,0.01000,3.207
157
+ 22,self_attn.q_proj,0.0000132499,0.01000,3.241
158
+ 22,self_attn.k_proj,0.0000125743,0.01000,3.242
159
+ 22,self_attn.o_proj,0.0001912692,0.01000,0.519
160
+ 22,mlp.gate_proj,0.0000432975,0.01000,1.459
161
+ 22,mlp.up_proj,0.0000557851,0.01000,1.467
162
+ 22,mlp.down_proj,0.0010292851,0.01000,2.225
163
+ 23,self_attn.q_proj,0.0000115864,0.01000,2.830
164
+ 23,self_attn.k_proj,0.0000120132,0.01000,2.833
165
+ 23,self_attn.v_proj,0.0000485714,0.01000,2.840
166
+ 23,self_attn.o_proj,0.0004808002,0.01000,0.514
167
+ 23,mlp.gate_proj,0.0000493452,0.01000,1.515
168
+ 23,mlp.up_proj,0.0000595773,0.01000,1.521
169
+ 23,mlp.down_proj,0.0022342433,0.01000,2.233
smollm2-1p7b__gptq__wikitext2__n512__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_49i1jpy7",
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__s1/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
smollm2-1p7b__gptq__wikitext2__n512__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
+ }