Upload folder using huggingface_hub

smollm2-1p7b__gptq__openhermes__n512__s1/BENCH_DONE

@@ -0,0 +1 @@
+2026-07-31T02:19:50.880296+00:00

smollm2-1p7b__gptq__openhermes__n512__s1/chat_template.jinja

@@ -0,0 +1,6 @@
+{% for message in messages %}{% if loop.first and messages[0]['role'] != 'system' %}{{ '<|im_start|>system
+You are a helpful AI assistant named SmolLM, trained by Hugging Face<|im_end|>
+' }}{% endif %}{{'<|im_start|>' + message['role'] + '
+' + message['content'] + '<|im_end|>' + '
+'}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant
+' }}{% endif %}

smollm2-1p7b__gptq__openhermes__n512__s1/config.json

@@ -0,0 +1,90 @@
+{
+  "architectures": [
+    "LlamaForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "bos_token_id": 1,
+  "dtype": "bfloat16",
+  "eos_token_id": 2,
+  "head_dim": 64,
+  "hidden_act": "silu",
+  "hidden_size": 2048,
+  "initializer_range": 0.02,
+  "intermediate_size": 8192,
+  "max_position_embeddings": 8192,
+  "mlp_bias": false,
+  "model_type": "llama",
+  "num_attention_heads": 32,
+  "num_hidden_layers": 24,
+  "num_key_value_heads": 32,
+  "pad_token_id": 2,
+  "pretraining_tp": 1,
+  "quantization_config": {
+    "bits": 4,
+    "checkpoint_format": "gptq",
+    "desc_act": true,
+    "format": "gptq",
+    "group_size": 128,
+    "lm_head": false,
+    "meta": {
+      "act_group_aware": false,
+      "auto_forward_data_parallel": true,
+      "damp_auto_increment": 0.01,
+      "damp_percent": 0.01,
+      "dense_vram_strategy": "exclusive",
+      "dense_vram_strategy_devices": null,
+      "fallback": {
+        "smooth": null,
+        "strategy": "rtn",
+        "threshold": "0.5%"
+      },
+      "foem": null,
+      "gc_mode": "interval",
+      "gptaq": null,
+      "hessian": {
+        "chunk_bytes": null,
+        "chunk_size": null,
+        "staging_dtype": "float32"
+      },
+      "mock_quantization": false,
+      "moe_vram_strategy": "exclusive",
+      "moe_vram_strategy_devices": null,
+      "mse": 0.0,
+      "offload_to_disk": true,
+      "offload_to_disk_path": "/tmp/gptqmodel_kigov_6f",
+      "pack_impl": "cpu",
+      "quantizer": [
+        "gptqmodel:7.3.2"
+      ],
+      "static_groups": false,
+      "true_sequential": true,
+      "uri": "https://github.com/modelcloud/gptqmodel",
+      "wait_for_submodule_finalizers": false
+    },
+    "method": "gptq",
+    "pack_dtype": "int32",
+    "quant_method": "gptq",
+    "sym": true
+  },
+  "rms_norm_eps": 1e-05,
+  "rope_parameters": {
+    "rope_theta": 130000,
+    "rope_type": "default"
+  },
+  "tie_word_embeddings": true,
+  "transformers.js_config": {
+    "dtype": "q4",
+    "kv_cache_dtype": {
+      "fp16": "float16",
+      "q4f16": "float16"
+    },
+    "use_external_data_format": {
+      "model.onnx": true,
+      "model_fp16.onnx": true
+    }
+  },
+  "transformers_version": "5.14.1",
+  "use_cache": true,
+  "vocab_size": 49152
+}

smollm2-1p7b__gptq__openhermes__n512__s1/generation_config.json

@@ -0,0 +1,8 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 1,
+  "do_sample": true,
+  "eos_token_id": 2,
+  "pad_token_id": 2,
+  "transformers_version": "5.14.1"
+}

smollm2-1p7b__gptq__openhermes__n512__s1/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c2599e0049f7cfce1643a2e2f5ef6cdbcd44761b952148f7664703a46e19c740
+size 1040336952

smollm2-1p7b__gptq__openhermes__n512__s1/quant_log.csv

@@ -0,0 +1,169 @@
+layer,module,loss,samples,damp,time
+0,self_attn.v_proj,0.0000000337,0.01000,2.477
+0,self_attn.k_proj,0.0000009402,0.01000,2.497
+0,self_attn.q_proj,0.0000012750,0.01000,2.506
+0,self_attn.o_proj,0.0000000277,0.01000,0.466
+0,mlp.up_proj,0.0000011604,0.01000,1.338
+0,mlp.gate_proj,0.0000012334,0.01000,1.341
+0,mlp.down_proj,0.0000006100,0.01000,2.095
+1,self_attn.v_proj,0.0000007157,0.01000,2.557
+1,self_attn.q_proj,0.0000036913,0.01000,2.572
+1,self_attn.k_proj,0.0000031694,0.01000,2.577
+1,self_attn.o_proj,0.0000003435,0.01000,0.476
+1,mlp.gate_proj,0.0000025351,0.01000,1.451
+1,mlp.up_proj,0.0000021889,0.01000,1.457
+1,mlp.down_proj,0.0000143261,0.01000,2.070
+2,self_attn.q_proj,0.0000059752,0.01000,2.572
+2,self_attn.k_proj,0.0000052081,0.01000,2.583
+2,self_attn.v_proj,0.0000020005,0.01000,2.588
+2,self_attn.o_proj,0.0000008835,0.01000,0.470
+2,mlp.up_proj,0.0000034365,0.01000,1.435
+2,mlp.gate_proj,0.0000038694,0.01000,1.442
+2,mlp.down_proj,0.0000025359,0.01000,2.119
+3,self_attn.k_proj,0.0000070476,0.01000,2.552
+3,self_attn.v_proj,0.0000028838,0.01000,2.567
+3,self_attn.q_proj,0.0000071103,0.01000,2.566
+3,self_attn.o_proj,0.0000011310,0.01000,0.476
+3,mlp.gate_proj,0.0000047080,0.01000,1.345
+3,mlp.up_proj,0.0000044615,0.01000,1.344
+3,mlp.down_proj,0.0000041197,0.01000,2.076
+4,self_attn.k_proj,0.0000075252,0.01000,2.574
+4,self_attn.q_proj,0.0000078007,0.01000,2.575
+4,self_attn.v_proj,0.0000031933,0.01000,2.584
+4,self_attn.o_proj,0.0000018478,0.01000,0.490
+4,mlp.up_proj,0.0000056542,0.01000,1.421
+4,mlp.gate_proj,0.0000058483,0.01000,1.429
+4,mlp.down_proj,0.0000054847,0.01000,2.134
+5,self_attn.v_proj,0.0000032445,0.01000,2.499
+5,self_attn.k_proj,0.0000071534,0.01000,2.516
+5,self_attn.q_proj,0.0000074921,0.01000,2.523
+5,self_attn.o_proj,0.0000021201,0.01000,0.472
+5,mlp.gate_proj,0.0000068453,0.01000,1.279
+5,mlp.up_proj,0.0000068738,0.01000,1.284
+5,mlp.down_proj,0.0000072885,0.01000,2.085
+6,self_attn.q_proj,0.0000087981,0.01000,2.500
+6,self_attn.v_proj,0.0000039846,0.01000,2.503
+6,self_attn.k_proj,0.0000081475,0.01000,2.513
+6,self_attn.o_proj,0.0000029795,0.01000,0.473
+6,mlp.gate_proj,0.0000082565,0.01000,1.284
+6,mlp.up_proj,0.0000084340,0.01000,1.285
+6,mlp.down_proj,0.0000103879,0.01000,2.078
+7,self_attn.k_proj,0.0000088426,0.01000,2.627
+7,self_attn.q_proj,0.0000112356,0.01000,2.631
+7,self_attn.v_proj,0.0000042538,0.01000,2.640
+7,self_attn.o_proj,0.0000045459,0.01000,0.474
+7,mlp.up_proj,0.0000107337,0.01000,1.362
+7,mlp.gate_proj,0.0000100838,0.01000,1.361
+7,mlp.down_proj,0.0000302919,0.01000,2.079
+8,self_attn.q_proj,0.0000085162,0.01000,2.914
+8,self_attn.k_proj,0.0000084858,0.01000,2.928
+8,self_attn.v_proj,0.0000057519,0.01000,2.937
+8,self_attn.o_proj,0.0000095754,0.01000,0.482
+8,mlp.gate_proj,0.0000118723,0.01000,1.362
+8,mlp.up_proj,0.0000124140,0.01000,1.366
+8,mlp.down_proj,0.0000208759,0.01000,2.106
+9,self_attn.v_proj,0.0000051524,0.01000,2.611
+9,self_attn.q_proj,0.0000081331,0.01000,2.629
+9,self_attn.k_proj,0.0000081119,0.01000,2.632
+9,self_attn.o_proj,0.0000107268,0.01000,0.486
+9,mlp.gate_proj,0.0000116126,0.01000,1.339
+9,mlp.up_proj,0.0000131634,0.01000,1.345
+9,mlp.down_proj,0.0000216112,0.01000,2.115
+10,self_attn.v_proj,0.0000052379,0.01000,2.790
+10,self_attn.k_proj,0.0000085315,0.01000,2.791
+10,self_attn.q_proj,0.0000087452,0.01000,2.803
+10,self_attn.o_proj,0.0000117420,0.01000,0.492
+10,mlp.up_proj,0.0000130506,0.01000,1.450
+10,mlp.gate_proj,0.0000117121,0.01000,1.469
+10,mlp.down_proj,0.0000258098,0.01000,2.083
+11,self_attn.k_proj,0.0000085197,0.01000,2.512
+11,self_attn.v_proj,0.0000051758,0.01000,2.520
+11,self_attn.q_proj,0.0000088408,0.01000,2.524
+11,self_attn.o_proj,0.0000140740,0.01000,0.471
+11,mlp.up_proj,0.0000131280,0.01000,1.295
+11,mlp.gate_proj,0.0000118507,0.01000,1.303
+11,mlp.down_proj,0.0000281570,0.01000,2.101
+12,self_attn.k_proj,0.0000093932,0.01000,2.524
+12,self_attn.v_proj,0.0000075252,0.01000,2.530
+12,self_attn.q_proj,0.0000100575,0.01000,2.540
+12,self_attn.o_proj,0.0000198036,0.01000,0.486
+12,mlp.up_proj,0.0000138866,0.01000,1.584
+12,mlp.gate_proj,0.0000124932,0.01000,1.601
+12,mlp.down_proj,0.0000331783,0.01000,2.104
+13,self_attn.v_proj,0.0000068582,0.01000,2.576
+13,self_attn.q_proj,0.0000100769,0.01000,2.590
+13,self_attn.k_proj,0.0000095404,0.01000,2.595
+13,self_attn.o_proj,0.0000169672,0.01000,0.475
+13,mlp.gate_proj,0.0000135795,0.01000,1.323
+13,mlp.up_proj,0.0000151426,0.01000,1.326
+13,mlp.down_proj,0.0000416107,0.01000,2.089
+14,self_attn.q_proj,0.0000095971,0.01000,2.741
+14,self_attn.k_proj,0.0000097609,0.01000,2.746
+14,self_attn.v_proj,0.0000075833,0.01000,2.751
+14,self_attn.o_proj,0.0000223590,0.01000,0.471
+14,mlp.gate_proj,0.0000132857,0.01000,1.298
+14,mlp.up_proj,0.0000151961,0.01000,1.300
+14,mlp.down_proj,0.0000499525,0.01000,2.077
+15,self_attn.k_proj,0.0000103901,0.01000,2.634
+15,self_attn.q_proj,0.0000106689,0.01000,2.640
+15,self_attn.v_proj,0.0000117326,0.01000,2.644
+15,self_attn.o_proj,0.0000295365,0.01000,0.480
+15,mlp.gate_proj,0.0000149296,0.01000,1.315
+15,mlp.up_proj,0.0000173615,0.01000,1.321
+15,mlp.down_proj,0.0000772218,0.01000,2.054
+16,self_attn.v_proj,0.0000111075,0.01000,2.423
+16,self_attn.k_proj,0.0000111738,0.01000,2.431
+16,self_attn.q_proj,0.0000112901,0.01000,2.437
+16,self_attn.o_proj,0.0000372591,0.01000,0.470
+16,mlp.gate_proj,0.0000186547,0.01000,1.346
+16,mlp.up_proj,0.0000220117,0.01000,1.350
+16,mlp.down_proj,0.0001235750,0.01000,2.072
+17,self_attn.v_proj,0.0000159544,0.01000,2.477
+17,self_attn.k_proj,0.0000110939,0.01000,2.486
+17,self_attn.q_proj,0.0000111173,0.01000,2.491
+17,self_attn.o_proj,0.0000627367,0.01000,0.488
+17,mlp.gate_proj,0.0000232565,0.01000,1.299
+17,mlp.up_proj,0.0000272129,0.01000,1.303
+17,mlp.down_proj,0.0001897762,0.01000,2.108
+18,self_attn.v_proj,0.0000189906,0.01000,2.524
+18,self_attn.q_proj,0.0000117331,0.01000,2.523
+18,self_attn.k_proj,0.0000112217,0.01000,2.537
+18,self_attn.o_proj,0.0000714254,0.01000,0.485
+18,mlp.gate_proj,0.0000294009,0.01000,1.266
+18,mlp.up_proj,0.0000356838,0.01000,1.268
+18,mlp.down_proj,0.0003198288,0.01000,2.078
+19,self_attn.k_proj,0.0000120458,0.01000,2.373
+19,self_attn.q_proj,0.0000126138,0.01000,2.386
+19,self_attn.v_proj,0.0000266571,0.01000,2.389
+19,self_attn.o_proj,0.0000707284,0.01000,0.469
+19,mlp.up_proj,0.0000426669,0.01000,1.272
+19,mlp.gate_proj,0.0000340257,0.01000,1.271
+19,mlp.down_proj,0.0004511232,0.01000,2.071
+20,self_attn.v_proj,0.0000315239,0.01000,2.406
+20,self_attn.k_proj,0.0000119271,0.01000,2.411
+20,self_attn.q_proj,0.0000127238,0.01000,2.418
+20,self_attn.o_proj,0.0000787687,0.01000,0.480
+20,mlp.up_proj,0.0000472725,0.01000,1.299
+20,mlp.gate_proj,0.0000371416,0.01000,1.310
+20,mlp.down_proj,0.0005506342,0.01000,2.062
+21,self_attn.k_proj,0.0000132218,0.01000,2.449
+21,self_attn.v_proj,0.0000394823,0.01000,2.448
+21,self_attn.q_proj,0.0000136714,0.01000,2.460
+21,self_attn.o_proj,0.0001064553,0.01000,0.474
+21,mlp.up_proj,0.0000514953,0.01000,1.385
+21,mlp.gate_proj,0.0000397390,0.01000,1.386
+21,mlp.down_proj,0.0006506395,0.01000,2.100
+22,self_attn.k_proj,0.0000130083,0.01000,2.752
+22,self_attn.v_proj,0.0000389843,0.01000,2.752
+22,self_attn.q_proj,0.0000136158,0.01000,2.766
+22,self_attn.o_proj,0.0001824538,0.01000,0.470
+22,mlp.gate_proj,0.0000430103,0.01000,1.503
+22,mlp.up_proj,0.0000552352,0.01000,1.508
+22,mlp.down_proj,0.0008570508,0.01000,2.077
+23,self_attn.q_proj,0.0000117033,0.01000,2.796
+23,self_attn.k_proj,0.0000121796,0.01000,2.805
+23,self_attn.v_proj,0.0000489267,0.01000,2.807
+23,self_attn.o_proj,0.0004572894,0.01000,0.480
+23,mlp.gate_proj,0.0000480475,0.01000,1.228
+23,mlp.up_proj,0.0000579260,0.01000,1.228
+23,mlp.down_proj,0.0017952052,0.01000,2.061

smollm2-1p7b__gptq__openhermes__n512__s1/quantize_config.json

@@ -0,0 +1,47 @@
+{
+  "bits": 4,
+  "group_size": 128,
+  "desc_act": true,
+  "lm_head": false,
+  "method": "gptq",
+  "quant_method": "gptq",
+  "format": "gptq",
+  "checkpoint_format": "gptq",
+  "pack_dtype": "int32",
+  "meta": {
+    "quantizer": [
+      "gptqmodel:7.3.2"
+    ],
+    "uri": "https://github.com/modelcloud/gptqmodel",
+    "damp_percent": 0.01,
+    "damp_auto_increment": 0.01,
+    "static_groups": false,
+    "true_sequential": true,
+    "mse": 0.0,
+    "gptaq": null,
+    "foem": null,
+    "act_group_aware": false,
+    "fallback": {
+      "strategy": "rtn",
+      "threshold": "0.5%",
+      "smooth": null
+    },
+    "offload_to_disk": true,
+    "offload_to_disk_path": "/tmp/gptqmodel_kigov_6f",
+    "pack_impl": "cpu",
+    "gc_mode": "interval",
+    "wait_for_submodule_finalizers": false,
+    "auto_forward_data_parallel": true,
+    "dense_vram_strategy": "exclusive",
+    "dense_vram_strategy_devices": null,
+    "moe_vram_strategy": "exclusive",
+    "moe_vram_strategy_devices": null,
+    "mock_quantization": false,
+    "hessian": {
+      "chunk_size": null,
+      "chunk_bytes": null,
+      "staging_dtype": "float32"
+    }
+  },
+  "sym": true
+}

smollm2-1p7b__gptq__openhermes__n512__s1/tokenizer_config.json

@@ -0,0 +1,19 @@
+{
+  "add_prefix_space": false,
+  "backend": "tokenizers",
+  "bos_token": "<|im_start|>",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "extra_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>"
+  ],
+  "is_local": false,
+  "local_files_only": false,
+  "model_max_length": 8192,
+  "pad_token": "<|im_end|>",
+  "tokenizer_class": "GPT2Tokenizer",
+  "unk_token": "<|endoftext|>",
+  "vocab_size": 49152
+}