Upload folder using huggingface_hub
Browse files- smollm2-1p7b__gptq__openhermes__n32__s0/BENCH_DONE +1 -0
- smollm2-1p7b__gptq__openhermes__n32__s0/chat_template.jinja +6 -0
- smollm2-1p7b__gptq__openhermes__n32__s0/config.json +90 -0
- smollm2-1p7b__gptq__openhermes__n32__s0/generation_config.json +8 -0
- smollm2-1p7b__gptq__openhermes__n32__s0/model.safetensors +3 -0
- smollm2-1p7b__gptq__openhermes__n32__s0/quant_log.csv +169 -0
- smollm2-1p7b__gptq__openhermes__n32__s0/quantize_config.json +47 -0
- smollm2-1p7b__gptq__openhermes__n32__s0/tokenizer.json +0 -0
- smollm2-1p7b__gptq__openhermes__n32__s0/tokenizer_config.json +19 -0
smollm2-1p7b__gptq__openhermes__n32__s0/BENCH_DONE
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
2026-07-31T01:21:22.860732+00:00
|
smollm2-1p7b__gptq__openhermes__n32__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__openhermes__n32__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_x_hj1lap",
|
| 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__openhermes__n32__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__openhermes__n32__s0/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4ae002d6dd220b7194492f5dee97370c711164e88eb8960a81e3f2331501b31c
|
| 3 |
+
size 1040336952
|
smollm2-1p7b__gptq__openhermes__n32__s0/quant_log.csv
ADDED
|
@@ -0,0 +1,169 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
layer,module,loss,samples,damp,time
|
| 2 |
+
0,self_attn.q_proj,0.0000184613,0.01000,3.475
|
| 3 |
+
0,self_attn.v_proj,0.0000005030,0.01000,2.923
|
| 4 |
+
0,self_attn.k_proj,0.0000137039,0.01000,2.957
|
| 5 |
+
0,self_attn.o_proj,0.0000004304,0.01000,0.448
|
| 6 |
+
0,mlp.up_proj,0.0000171527,0.01000,1.361
|
| 7 |
+
0,mlp.gate_proj,0.0000182089,0.01000,1.365
|
| 8 |
+
0,mlp.down_proj,0.0000072308,0.01000,2.006
|
| 9 |
+
1,self_attn.q_proj,0.0000552167,0.01000,3.217
|
| 10 |
+
1,self_attn.v_proj,0.0000107853,0.01000,3.227
|
| 11 |
+
1,self_attn.k_proj,0.0000474163,0.01000,3.237
|
| 12 |
+
1,self_attn.o_proj,0.0000049679,0.01000,0.457
|
| 13 |
+
1,mlp.gate_proj,0.0000382429,0.01000,1.368
|
| 14 |
+
1,mlp.up_proj,0.0000330544,0.01000,1.367
|
| 15 |
+
1,mlp.down_proj,0.0002232692,0.01000,2.065
|
| 16 |
+
2,self_attn.k_proj,0.0000783847,0.01000,3.062
|
| 17 |
+
2,self_attn.q_proj,0.0000891805,0.01000,3.062
|
| 18 |
+
2,self_attn.v_proj,0.0000301912,0.01000,3.064
|
| 19 |
+
2,self_attn.o_proj,0.0000127110,0.01000,0.456
|
| 20 |
+
2,mlp.up_proj,0.0000515723,0.01000,1.577
|
| 21 |
+
2,mlp.gate_proj,0.0000580461,0.01000,1.586
|
| 22 |
+
2,mlp.down_proj,0.0000323391,0.01000,1.999
|
| 23 |
+
3,self_attn.q_proj,0.0001054438,0.01000,3.032
|
| 24 |
+
3,self_attn.k_proj,0.0001048935,0.01000,3.034
|
| 25 |
+
3,self_attn.v_proj,0.0000432484,0.01000,3.042
|
| 26 |
+
3,self_attn.o_proj,0.0000162795,0.01000,0.469
|
| 27 |
+
3,mlp.up_proj,0.0000664950,0.01000,1.423
|
| 28 |
+
3,mlp.gate_proj,0.0000700708,0.01000,1.432
|
| 29 |
+
3,mlp.down_proj,0.0000519395,0.01000,2.065
|
| 30 |
+
4,self_attn.k_proj,0.0001111413,0.01000,2.690
|
| 31 |
+
4,self_attn.q_proj,0.0001151028,0.01000,2.704
|
| 32 |
+
4,self_attn.v_proj,0.0000473989,0.01000,2.705
|
| 33 |
+
4,self_attn.o_proj,0.0000263754,0.01000,0.472
|
| 34 |
+
4,mlp.up_proj,0.0000840555,0.01000,1.348
|
| 35 |
+
4,mlp.gate_proj,0.0000869420,0.01000,1.350
|
| 36 |
+
4,mlp.down_proj,0.0000706694,0.01000,2.042
|
| 37 |
+
5,self_attn.v_proj,0.0000484096,0.01000,2.640
|
| 38 |
+
5,self_attn.k_proj,0.0001061184,0.01000,2.645
|
| 39 |
+
5,self_attn.q_proj,0.0001109536,0.01000,2.658
|
| 40 |
+
5,self_attn.o_proj,0.0000300812,0.01000,0.465
|
| 41 |
+
5,mlp.gate_proj,0.0001025333,0.01000,1.396
|
| 42 |
+
5,mlp.up_proj,0.0001030941,0.01000,1.401
|
| 43 |
+
5,mlp.down_proj,0.0000960554,0.01000,2.054
|
| 44 |
+
6,self_attn.k_proj,0.0001222574,0.01000,2.674
|
| 45 |
+
6,self_attn.q_proj,0.0001313874,0.01000,2.685
|
| 46 |
+
6,self_attn.v_proj,0.0000599564,0.01000,2.691
|
| 47 |
+
6,self_attn.o_proj,0.0000435177,0.01000,0.466
|
| 48 |
+
6,mlp.up_proj,0.0001271102,0.01000,1.396
|
| 49 |
+
6,mlp.gate_proj,0.0001242665,0.01000,1.400
|
| 50 |
+
6,mlp.down_proj,0.0001394397,0.01000,2.054
|
| 51 |
+
7,self_attn.k_proj,0.0001328580,0.01000,2.758
|
| 52 |
+
7,self_attn.q_proj,0.0001676148,0.01000,2.768
|
| 53 |
+
7,self_attn.v_proj,0.0000644012,0.01000,2.774
|
| 54 |
+
7,self_attn.o_proj,0.0000668362,0.01000,0.466
|
| 55 |
+
7,mlp.gate_proj,0.0001519495,0.01000,1.358
|
| 56 |
+
7,mlp.up_proj,0.0001618967,0.01000,1.360
|
| 57 |
+
7,mlp.down_proj,0.0004434982,0.01000,2.050
|
| 58 |
+
8,self_attn.v_proj,0.0000873090,0.01000,2.719
|
| 59 |
+
8,self_attn.k_proj,0.0001277085,0.01000,2.725
|
| 60 |
+
8,self_attn.q_proj,0.0001287549,0.01000,2.733
|
| 61 |
+
8,self_attn.o_proj,0.0001346973,0.01000,0.468
|
| 62 |
+
8,mlp.gate_proj,0.0001774206,0.01000,1.495
|
| 63 |
+
8,mlp.up_proj,0.0001858333,0.01000,1.497
|
| 64 |
+
8,mlp.down_proj,0.0002705044,0.01000,2.067
|
| 65 |
+
9,self_attn.v_proj,0.0000770806,0.01000,2.702
|
| 66 |
+
9,self_attn.k_proj,0.0001207331,0.01000,2.711
|
| 67 |
+
9,self_attn.q_proj,0.0001211217,0.01000,2.720
|
| 68 |
+
9,self_attn.o_proj,0.0001547982,0.01000,0.476
|
| 69 |
+
9,mlp.up_proj,0.0001962209,0.01000,1.363
|
| 70 |
+
9,mlp.gate_proj,0.0001729137,0.01000,1.367
|
| 71 |
+
9,mlp.down_proj,0.0002874191,0.01000,2.044
|
| 72 |
+
10,self_attn.v_proj,0.0000788475,0.01000,2.694
|
| 73 |
+
10,self_attn.q_proj,0.0001309472,0.01000,2.710
|
| 74 |
+
10,self_attn.k_proj,0.0001275659,0.01000,2.712
|
| 75 |
+
10,self_attn.o_proj,0.0001678705,0.01000,0.476
|
| 76 |
+
10,mlp.gate_proj,0.0001745926,0.01000,1.526
|
| 77 |
+
10,mlp.up_proj,0.0001949321,0.01000,1.539
|
| 78 |
+
10,mlp.down_proj,0.0003484071,0.01000,2.098
|
| 79 |
+
11,self_attn.v_proj,0.0000780905,0.01000,2.746
|
| 80 |
+
11,self_attn.k_proj,0.0001277797,0.01000,2.760
|
| 81 |
+
11,self_attn.q_proj,0.0001326126,0.01000,2.765
|
| 82 |
+
11,self_attn.o_proj,0.0001997200,0.01000,0.473
|
| 83 |
+
11,mlp.gate_proj,0.0001760165,0.01000,1.433
|
| 84 |
+
11,mlp.up_proj,0.0001952612,0.01000,1.435
|
| 85 |
+
11,mlp.down_proj,0.0003800914,0.01000,2.055
|
| 86 |
+
12,self_attn.q_proj,0.0001499186,0.01000,2.719
|
| 87 |
+
12,self_attn.v_proj,0.0001126376,0.01000,2.728
|
| 88 |
+
12,self_attn.k_proj,0.0001398947,0.01000,2.736
|
| 89 |
+
12,self_attn.o_proj,0.0002767842,0.01000,0.472
|
| 90 |
+
12,mlp.up_proj,0.0002063411,0.01000,1.511
|
| 91 |
+
12,mlp.gate_proj,0.0001852478,0.01000,1.530
|
| 92 |
+
12,mlp.down_proj,0.0004446877,0.01000,2.072
|
| 93 |
+
13,self_attn.v_proj,0.0001029299,0.01000,2.760
|
| 94 |
+
13,self_attn.k_proj,0.0001420172,0.01000,2.769
|
| 95 |
+
13,self_attn.q_proj,0.0001498544,0.01000,2.778
|
| 96 |
+
13,self_attn.o_proj,0.0002353157,0.01000,0.471
|
| 97 |
+
13,mlp.gate_proj,0.0002036395,0.01000,1.450
|
| 98 |
+
13,mlp.up_proj,0.0002272346,0.01000,1.455
|
| 99 |
+
13,mlp.down_proj,0.0005714700,0.01000,2.076
|
| 100 |
+
14,self_attn.q_proj,0.0001452633,0.01000,2.916
|
| 101 |
+
14,self_attn.v_proj,0.0001149308,0.01000,2.929
|
| 102 |
+
14,self_attn.k_proj,0.0001471024,0.01000,2.937
|
| 103 |
+
14,self_attn.o_proj,0.0003250685,0.01000,0.475
|
| 104 |
+
14,mlp.up_proj,0.0002272049,0.01000,1.593
|
| 105 |
+
14,mlp.gate_proj,0.0001983413,0.01000,1.614
|
| 106 |
+
14,mlp.down_proj,0.0006897425,0.01000,2.068
|
| 107 |
+
15,self_attn.q_proj,0.0001598870,0.01000,2.739
|
| 108 |
+
15,self_attn.k_proj,0.0001553172,0.01000,2.754
|
| 109 |
+
15,self_attn.v_proj,0.0001769631,0.01000,2.759
|
| 110 |
+
15,self_attn.o_proj,0.0004033549,0.01000,0.478
|
| 111 |
+
15,mlp.up_proj,0.0002600948,0.01000,1.509
|
| 112 |
+
15,mlp.gate_proj,0.0002229825,0.01000,1.517
|
| 113 |
+
15,mlp.down_proj,0.0010481894,0.01000,2.086
|
| 114 |
+
16,self_attn.q_proj,0.0001686935,0.01000,2.861
|
| 115 |
+
16,self_attn.k_proj,0.0001663259,0.01000,2.873
|
| 116 |
+
16,self_attn.v_proj,0.0001668895,0.01000,2.876
|
| 117 |
+
16,self_attn.o_proj,0.0005331542,0.01000,0.488
|
| 118 |
+
16,mlp.gate_proj,0.0002787018,0.01000,1.629
|
| 119 |
+
16,mlp.up_proj,0.0003297016,0.01000,1.637
|
| 120 |
+
16,mlp.down_proj,0.0016766530,0.01000,2.105
|
| 121 |
+
17,self_attn.v_proj,0.0002383069,0.01000,2.918
|
| 122 |
+
17,self_attn.k_proj,0.0001656936,0.01000,2.927
|
| 123 |
+
17,self_attn.q_proj,0.0001670973,0.01000,2.937
|
| 124 |
+
17,self_attn.o_proj,0.0009057788,0.01000,0.486
|
| 125 |
+
17,mlp.gate_proj,0.0003478720,0.01000,1.583
|
| 126 |
+
17,mlp.up_proj,0.0004080457,0.01000,1.588
|
| 127 |
+
17,mlp.down_proj,0.0025944277,0.01000,2.111
|
| 128 |
+
18,self_attn.k_proj,0.0001672320,0.01000,2.915
|
| 129 |
+
18,self_attn.v_proj,0.0002838329,0.01000,2.917
|
| 130 |
+
18,self_attn.q_proj,0.0001750624,0.01000,2.924
|
| 131 |
+
18,self_attn.o_proj,0.0010082785,0.01000,0.484
|
| 132 |
+
18,mlp.gate_proj,0.0004420782,0.01000,1.566
|
| 133 |
+
18,mlp.up_proj,0.0005376945,0.01000,1.579
|
| 134 |
+
18,mlp.down_proj,0.0043666116,0.01000,2.110
|
| 135 |
+
19,self_attn.v_proj,0.0003979214,0.01000,3.069
|
| 136 |
+
19,self_attn.q_proj,0.0001887794,0.01000,3.076
|
| 137 |
+
19,self_attn.k_proj,0.0001796075,0.01000,3.085
|
| 138 |
+
19,self_attn.o_proj,0.0009920824,0.01000,0.482
|
| 139 |
+
19,mlp.up_proj,0.0006397703,0.01000,1.560
|
| 140 |
+
19,mlp.gate_proj,0.0005093174,0.01000,1.566
|
| 141 |
+
19,mlp.down_proj,0.0060390416,0.01000,2.109
|
| 142 |
+
20,self_attn.v_proj,0.0004682927,0.01000,3.267
|
| 143 |
+
20,self_attn.q_proj,0.0001886797,0.01000,3.282
|
| 144 |
+
20,self_attn.k_proj,0.0001772726,0.01000,3.284
|
| 145 |
+
20,self_attn.o_proj,0.0010782894,0.01000,0.487
|
| 146 |
+
20,mlp.up_proj,0.0007071099,0.01000,1.512
|
| 147 |
+
20,mlp.gate_proj,0.0005543453,0.01000,1.520
|
| 148 |
+
20,mlp.down_proj,0.0071963943,0.01000,2.095
|
| 149 |
+
21,self_attn.k_proj,0.0001953098,0.01000,3.184
|
| 150 |
+
21,self_attn.q_proj,0.0002028941,0.01000,3.195
|
| 151 |
+
21,self_attn.v_proj,0.0005859086,0.01000,3.200
|
| 152 |
+
21,self_attn.o_proj,0.0014764997,0.01000,0.487
|
| 153 |
+
21,mlp.gate_proj,0.0005960375,0.01000,1.680
|
| 154 |
+
21,mlp.up_proj,0.0007730889,0.01000,1.697
|
| 155 |
+
21,mlp.down_proj,0.0085041728,0.01000,2.147
|
| 156 |
+
22,self_attn.k_proj,0.0001934035,0.01000,3.234
|
| 157 |
+
22,self_attn.v_proj,0.0005835084,0.01000,3.245
|
| 158 |
+
22,self_attn.q_proj,0.0002028819,0.01000,3.260
|
| 159 |
+
22,self_attn.o_proj,0.0024981603,0.01000,0.484
|
| 160 |
+
22,mlp.gate_proj,0.0006450476,0.01000,1.567
|
| 161 |
+
22,mlp.up_proj,0.0008300739,0.01000,1.579
|
| 162 |
+
22,mlp.down_proj,0.0117292469,0.01000,2.076
|
| 163 |
+
23,self_attn.k_proj,0.0001806692,0.01000,2.779
|
| 164 |
+
23,self_attn.q_proj,0.0001741704,0.01000,2.790
|
| 165 |
+
23,self_attn.v_proj,0.0007296070,0.01000,2.799
|
| 166 |
+
23,self_attn.o_proj,0.0070117684,0.01000,0.474
|
| 167 |
+
23,mlp.up_proj,0.0008680237,0.01000,1.487
|
| 168 |
+
23,mlp.gate_proj,0.0007188843,0.01000,1.492
|
| 169 |
+
23,mlp.down_proj,0.0237775221,0.01000,2.046
|
smollm2-1p7b__gptq__openhermes__n32__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_x_hj1lap",
|
| 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__openhermes__n32__s0/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
smollm2-1p7b__gptq__openhermes__n32__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 |
+
}
|