Upload folder using huggingface_hub
Browse files- smollm2-1p7b__gptq__wikitext2__n128__s0/BENCH_DONE +1 -0
- smollm2-1p7b__gptq__wikitext2__n128__s0/chat_template.jinja +6 -0
- smollm2-1p7b__gptq__wikitext2__n128__s0/config.json +90 -0
- smollm2-1p7b__gptq__wikitext2__n128__s0/generation_config.json +8 -0
- smollm2-1p7b__gptq__wikitext2__n128__s0/model.safetensors +3 -0
- smollm2-1p7b__gptq__wikitext2__n128__s0/quant_log.csv +169 -0
- smollm2-1p7b__gptq__wikitext2__n128__s0/quantize_config.json +47 -0
- smollm2-1p7b__gptq__wikitext2__n128__s0/tokenizer.json +0 -0
- smollm2-1p7b__gptq__wikitext2__n128__s0/tokenizer_config.json +19 -0
smollm2-1p7b__gptq__wikitext2__n128__s0/BENCH_DONE
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
2026-07-31T00:15:24.942589+00:00
|
smollm2-1p7b__gptq__wikitext2__n128__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__n128__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_foaanks6",
|
| 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__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__n128__s0/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:48d288358e3ae6493150b95cc8ad383dadde05b68e1cc981981503db09d22b4b
|
| 3 |
+
size 1040336952
|
smollm2-1p7b__gptq__wikitext2__n128__s0/quant_log.csv
ADDED
|
@@ -0,0 +1,169 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
layer,module,loss,samples,damp,time
|
| 2 |
+
0,self_attn.k_proj,0.0000036832,0.01000,2.721
|
| 3 |
+
0,self_attn.q_proj,0.0000050271,0.01000,2.749
|
| 4 |
+
0,self_attn.v_proj,0.0000001309,0.01000,2.747
|
| 5 |
+
0,self_attn.o_proj,0.0000001100,0.01000,0.514
|
| 6 |
+
0,mlp.gate_proj,0.0000054342,0.01000,1.507
|
| 7 |
+
0,mlp.up_proj,0.0000051175,0.01000,1.518
|
| 8 |
+
0,mlp.down_proj,0.0000028524,0.01000,2.223
|
| 9 |
+
1,self_attn.q_proj,0.0000138584,0.01000,2.749
|
| 10 |
+
1,self_attn.v_proj,0.0000029816,0.01000,2.761
|
| 11 |
+
1,self_attn.k_proj,0.0000124204,0.01000,2.765
|
| 12 |
+
1,self_attn.o_proj,0.0000012341,0.01000,0.521
|
| 13 |
+
1,mlp.gate_proj,0.0000103070,0.01000,1.489
|
| 14 |
+
1,mlp.up_proj,0.0000089078,0.01000,1.496
|
| 15 |
+
1,mlp.down_proj,0.0000468853,0.01000,2.247
|
| 16 |
+
2,self_attn.k_proj,0.0000203927,0.01000,2.940
|
| 17 |
+
2,self_attn.q_proj,0.0000227895,0.01000,2.967
|
| 18 |
+
2,self_attn.v_proj,0.0000079711,0.01000,2.971
|
| 19 |
+
2,self_attn.o_proj,0.0000031613,0.01000,0.525
|
| 20 |
+
2,mlp.up_proj,0.0000142304,0.01000,1.678
|
| 21 |
+
2,mlp.gate_proj,0.0000159906,0.01000,1.686
|
| 22 |
+
2,mlp.down_proj,0.0000118487,0.01000,2.253
|
| 23 |
+
3,self_attn.k_proj,0.0000281952,0.01000,2.921
|
| 24 |
+
3,self_attn.v_proj,0.0000118035,0.01000,2.939
|
| 25 |
+
3,self_attn.q_proj,0.0000281271,0.01000,2.941
|
| 26 |
+
3,self_attn.o_proj,0.0000040222,0.01000,0.514
|
| 27 |
+
3,mlp.gate_proj,0.0000202460,0.01000,1.820
|
| 28 |
+
3,mlp.up_proj,0.0000192468,0.01000,1.831
|
| 29 |
+
3,mlp.down_proj,0.0000192887,0.01000,2.272
|
| 30 |
+
4,self_attn.v_proj,0.0000133510,0.01000,2.942
|
| 31 |
+
4,self_attn.q_proj,0.0000315981,0.01000,2.962
|
| 32 |
+
4,self_attn.k_proj,0.0000305773,0.01000,2.968
|
| 33 |
+
4,self_attn.o_proj,0.0000067626,0.01000,0.517
|
| 34 |
+
4,mlp.gate_proj,0.0000252795,0.01000,1.845
|
| 35 |
+
4,mlp.up_proj,0.0000244634,0.01000,1.853
|
| 36 |
+
4,mlp.down_proj,0.0000274926,0.01000,2.252
|
| 37 |
+
5,self_attn.q_proj,0.0000307069,0.01000,2.874
|
| 38 |
+
5,self_attn.v_proj,0.0000136365,0.01000,2.888
|
| 39 |
+
5,self_attn.k_proj,0.0000295990,0.01000,2.897
|
| 40 |
+
5,self_attn.o_proj,0.0000086795,0.01000,0.516
|
| 41 |
+
5,mlp.gate_proj,0.0000299213,0.01000,1.505
|
| 42 |
+
5,mlp.up_proj,0.0000300729,0.01000,1.509
|
| 43 |
+
5,mlp.down_proj,0.0000355307,0.01000,2.255
|
| 44 |
+
6,self_attn.k_proj,0.0000340926,0.01000,3.131
|
| 45 |
+
6,self_attn.q_proj,0.0000362841,0.01000,3.157
|
| 46 |
+
6,self_attn.v_proj,0.0000168440,0.01000,3.161
|
| 47 |
+
6,self_attn.o_proj,0.0000108337,0.01000,0.522
|
| 48 |
+
6,mlp.gate_proj,0.0000355665,0.01000,1.775
|
| 49 |
+
6,mlp.up_proj,0.0000364002,0.01000,1.787
|
| 50 |
+
6,mlp.down_proj,0.0000479373,0.01000,2.233
|
| 51 |
+
7,self_attn.k_proj,0.0000352083,0.01000,3.036
|
| 52 |
+
7,self_attn.q_proj,0.0000435974,0.01000,3.057
|
| 53 |
+
7,self_attn.v_proj,0.0000172921,0.01000,3.057
|
| 54 |
+
7,self_attn.o_proj,0.0000176916,0.01000,0.532
|
| 55 |
+
7,mlp.up_proj,0.0000442982,0.01000,1.606
|
| 56 |
+
7,mlp.gate_proj,0.0000415601,0.01000,1.616
|
| 57 |
+
7,mlp.down_proj,0.0001203577,0.01000,2.258
|
| 58 |
+
8,self_attn.k_proj,0.0000333230,0.01000,2.845
|
| 59 |
+
8,self_attn.v_proj,0.0000232107,0.01000,2.855
|
| 60 |
+
8,self_attn.q_proj,0.0000337105,0.01000,2.867
|
| 61 |
+
8,self_attn.o_proj,0.0000306842,0.01000,0.523
|
| 62 |
+
8,mlp.gate_proj,0.0000487031,0.01000,1.741
|
| 63 |
+
8,mlp.up_proj,0.0000509861,0.01000,1.751
|
| 64 |
+
8,mlp.down_proj,0.0000866841,0.01000,2.267
|
| 65 |
+
9,self_attn.q_proj,0.0000312561,0.01000,2.961
|
| 66 |
+
9,self_attn.v_proj,0.0000200157,0.01000,2.979
|
| 67 |
+
9,self_attn.k_proj,0.0000310928,0.01000,2.982
|
| 68 |
+
9,self_attn.o_proj,0.0000422266,0.01000,0.517
|
| 69 |
+
9,mlp.gate_proj,0.0000461316,0.01000,1.545
|
| 70 |
+
9,mlp.up_proj,0.0000523846,0.01000,1.560
|
| 71 |
+
9,mlp.down_proj,0.0000904271,0.01000,2.248
|
| 72 |
+
10,self_attn.v_proj,0.0000205565,0.01000,2.723
|
| 73 |
+
10,self_attn.k_proj,0.0000329367,0.01000,2.732
|
| 74 |
+
10,self_attn.q_proj,0.0000337023,0.01000,2.749
|
| 75 |
+
10,self_attn.o_proj,0.0000518111,0.01000,0.514
|
| 76 |
+
10,mlp.gate_proj,0.0000454544,0.01000,1.546
|
| 77 |
+
10,mlp.up_proj,0.0000509552,0.01000,1.558
|
| 78 |
+
10,mlp.down_proj,0.0001051230,0.01000,2.244
|
| 79 |
+
11,self_attn.k_proj,0.0000327909,0.01000,2.783
|
| 80 |
+
11,self_attn.v_proj,0.0000201714,0.01000,2.801
|
| 81 |
+
11,self_attn.q_proj,0.0000338828,0.01000,2.809
|
| 82 |
+
11,self_attn.o_proj,0.0000560094,0.01000,0.514
|
| 83 |
+
11,mlp.up_proj,0.0000500892,0.01000,1.575
|
| 84 |
+
11,mlp.gate_proj,0.0000450670,0.01000,1.598
|
| 85 |
+
11,mlp.down_proj,0.0001100503,0.01000,2.252
|
| 86 |
+
12,self_attn.q_proj,0.0000381675,0.01000,2.778
|
| 87 |
+
12,self_attn.k_proj,0.0000358110,0.01000,2.776
|
| 88 |
+
12,self_attn.v_proj,0.0000287548,0.01000,2.786
|
| 89 |
+
12,self_attn.o_proj,0.0000733939,0.01000,0.521
|
| 90 |
+
12,mlp.up_proj,0.0000524407,0.01000,1.561
|
| 91 |
+
12,mlp.gate_proj,0.0000470782,0.01000,1.563
|
| 92 |
+
12,mlp.down_proj,0.0001212832,0.01000,2.250
|
| 93 |
+
13,self_attn.k_proj,0.0000349899,0.01000,2.858
|
| 94 |
+
13,self_attn.v_proj,0.0000253665,0.01000,2.881
|
| 95 |
+
13,self_attn.q_proj,0.0000365244,0.01000,2.885
|
| 96 |
+
13,self_attn.o_proj,0.0000632957,0.01000,0.519
|
| 97 |
+
13,mlp.gate_proj,0.0000478071,0.01000,1.489
|
| 98 |
+
13,mlp.up_proj,0.0000537333,0.01000,1.498
|
| 99 |
+
13,mlp.down_proj,0.0001389795,0.01000,2.255
|
| 100 |
+
14,self_attn.v_proj,0.0000272931,0.01000,2.817
|
| 101 |
+
14,self_attn.k_proj,0.0000348568,0.01000,2.833
|
| 102 |
+
14,self_attn.q_proj,0.0000344426,0.01000,2.840
|
| 103 |
+
14,self_attn.o_proj,0.0000798368,0.01000,0.521
|
| 104 |
+
14,mlp.up_proj,0.0000529273,0.01000,1.436
|
| 105 |
+
14,mlp.gate_proj,0.0000460209,0.01000,1.435
|
| 106 |
+
14,mlp.down_proj,0.0001672932,0.01000,2.231
|
| 107 |
+
15,self_attn.q_proj,0.0000379335,0.01000,2.895
|
| 108 |
+
15,self_attn.v_proj,0.0000419168,0.01000,2.895
|
| 109 |
+
15,self_attn.k_proj,0.0000368447,0.01000,2.898
|
| 110 |
+
15,self_attn.o_proj,0.0000918975,0.01000,0.523
|
| 111 |
+
15,mlp.gate_proj,0.0000492750,0.01000,1.656
|
| 112 |
+
15,mlp.up_proj,0.0000574582,0.01000,1.661
|
| 113 |
+
15,mlp.down_proj,0.0002396689,0.01000,2.258
|
| 114 |
+
16,self_attn.k_proj,0.0000371662,0.01000,2.821
|
| 115 |
+
16,self_attn.v_proj,0.0000374134,0.01000,2.830
|
| 116 |
+
16,self_attn.q_proj,0.0000373202,0.01000,2.842
|
| 117 |
+
16,self_attn.o_proj,0.0001106395,0.01000,0.521
|
| 118 |
+
16,mlp.up_proj,0.0000707617,0.01000,1.615
|
| 119 |
+
16,mlp.gate_proj,0.0000598458,0.01000,1.627
|
| 120 |
+
16,mlp.down_proj,0.0003891182,0.01000,2.312
|
| 121 |
+
17,self_attn.k_proj,0.0000366290,0.01000,3.435
|
| 122 |
+
17,self_attn.v_proj,0.0000526797,0.01000,3.469
|
| 123 |
+
17,self_attn.q_proj,0.0000369264,0.01000,3.493
|
| 124 |
+
17,self_attn.o_proj,0.0001843320,0.01000,0.543
|
| 125 |
+
17,mlp.up_proj,0.0000880967,0.01000,1.605
|
| 126 |
+
17,mlp.gate_proj,0.0000750517,0.01000,1.635
|
| 127 |
+
17,mlp.down_proj,0.0006181989,0.01000,2.276
|
| 128 |
+
18,self_attn.k_proj,0.0000376393,0.01000,2.966
|
| 129 |
+
18,self_attn.q_proj,0.0000392947,0.01000,2.974
|
| 130 |
+
18,self_attn.v_proj,0.0000639098,0.01000,2.994
|
| 131 |
+
18,self_attn.o_proj,0.0002262830,0.01000,0.545
|
| 132 |
+
18,mlp.gate_proj,0.0000958010,0.01000,1.663
|
| 133 |
+
18,mlp.up_proj,0.0001165359,0.01000,1.678
|
| 134 |
+
18,mlp.down_proj,0.0010496113,0.01000,2.273
|
| 135 |
+
19,self_attn.v_proj,0.0000913957,0.01000,2.949
|
| 136 |
+
19,self_attn.q_proj,0.0000434758,0.01000,2.964
|
| 137 |
+
19,self_attn.k_proj,0.0000412106,0.01000,2.971
|
| 138 |
+
19,self_attn.o_proj,0.0002683001,0.01000,0.519
|
| 139 |
+
19,mlp.up_proj,0.0001414062,0.01000,1.575
|
| 140 |
+
19,mlp.gate_proj,0.0001125368,0.01000,1.593
|
| 141 |
+
19,mlp.down_proj,0.0016581940,0.01000,2.258
|
| 142 |
+
20,self_attn.q_proj,0.0000441593,0.01000,2.912
|
| 143 |
+
20,self_attn.k_proj,0.0000413342,0.01000,2.922
|
| 144 |
+
20,self_attn.v_proj,0.0001096015,0.01000,2.922
|
| 145 |
+
20,self_attn.o_proj,0.0002804546,0.01000,0.513
|
| 146 |
+
20,mlp.up_proj,0.0001650621,0.01000,1.495
|
| 147 |
+
20,mlp.gate_proj,0.0001293615,0.01000,1.512
|
| 148 |
+
20,mlp.down_proj,0.0022521478,0.01000,2.253
|
| 149 |
+
21,self_attn.q_proj,0.0000496657,0.01000,2.782
|
| 150 |
+
21,self_attn.v_proj,0.0001434292,0.01000,2.782
|
| 151 |
+
21,self_attn.k_proj,0.0000476855,0.01000,2.789
|
| 152 |
+
21,self_attn.o_proj,0.0004433230,0.01000,0.532
|
| 153 |
+
21,mlp.gate_proj,0.0001462612,0.01000,1.572
|
| 154 |
+
21,mlp.up_proj,0.0001899343,0.01000,1.573
|
| 155 |
+
21,mlp.down_proj,0.0029545759,0.01000,2.269
|
| 156 |
+
22,self_attn.v_proj,0.0001481792,0.01000,2.796
|
| 157 |
+
22,self_attn.k_proj,0.0000487495,0.01000,2.807
|
| 158 |
+
22,self_attn.q_proj,0.0000513461,0.01000,2.812
|
| 159 |
+
22,self_attn.o_proj,0.0007236383,0.01000,0.523
|
| 160 |
+
22,mlp.up_proj,0.0002176035,0.01000,1.476
|
| 161 |
+
22,mlp.gate_proj,0.0001687118,0.01000,1.481
|
| 162 |
+
22,mlp.down_proj,0.0038872103,0.01000,2.255
|
| 163 |
+
23,self_attn.q_proj,0.0000451282,0.01000,2.868
|
| 164 |
+
23,self_attn.v_proj,0.0001890724,0.01000,2.889
|
| 165 |
+
23,self_attn.k_proj,0.0000466227,0.01000,2.896
|
| 166 |
+
23,self_attn.o_proj,0.0018031199,0.01000,0.520
|
| 167 |
+
23,mlp.up_proj,0.0002330264,0.01000,1.491
|
| 168 |
+
23,mlp.gate_proj,0.0001930177,0.01000,1.499
|
| 169 |
+
23,mlp.down_proj,0.0084164692,0.01000,2.235
|
smollm2-1p7b__gptq__wikitext2__n128__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_foaanks6",
|
| 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__s0/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
smollm2-1p7b__gptq__wikitext2__n128__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 |
+
}
|