Upload folder using huggingface_hub

config.json

@@ -32,16 +32,15 @@
             "num_bits": 8,
             "observer": "minmax",
             "observer_kwargs": {},
-            "strategy": "tensor",
+            "strategy": "channel",
             "symmetric": true,
             "type": "int"
           }
         }
       },
       "format": "int-quantized",
-      "global_compression_ratio": 1.4051191328987915,
+      "global_compression_ratio": 1.2390773684863086,
       "ignore": [
-        "model.layers.0.mlp.down_proj",
         "lm_head"
       ],
       "quant_method": "sparseml",
@@ -49,9 +48,9 @@
     },
     "sparsity_config": {
       "format": "dense",
-      "global_sparsity": 7.700428565876607,
+      "global_sparsity": 1.188444907528722,
       "registry_requires_subclass": false,
-      "sparsity_structure": "0:0"
+      "sparsity_structure": "unstructured"
     }
   },
   "eos_token_id": 2,
@@ -69,8 +68,8 @@
   "rope_scaling": null,
   "rope_theta": 10000.0,
   "tie_word_embeddings": false,
-  "torch_dtype": "float32",
-  "transformers_version": "4.39.3",
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.40.0",
   "use_cache": true,
   "vocab_size": 32000
 }

generation_config.json

@@ -3,5 +3,5 @@
   "eos_token_id": 2,
   "max_length": 2048,
   "pad_token_id": 0,
-  "transformers_version": "4.39.3"
+  "transformers_version": "4.40.0"
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f1413b4a713469abe8b3261fa9aa36d54712fdaa94ef5be5f0ab71b435714735
-size 1528201733
+oid sha256:ed7796316a087eaf71963e35ff79ae097d02000411fc834b315d0bf399ad870c
+size 1232041608

recipe.yaml

@@ -1,16 +1,10 @@
-test_stage:
+quant_stage:
   quant_modifiers:
-    vLLMQuantizationModifier:
-      ignore: [lm_head, model.layers.0.mlp.down_proj]
+    GPTQModifier:
+      sequential_update: false
+      ignore: [lm_head]
       config_groups:
         group_0:
-          weights: {num_bits: 8, type: int, symmetric: true, strategy: tensor}
-          input_activations: {num_bits: 8, type: int, symmetric: true, strategy: token, dynamic: true}
-          output_activations: null
+          weights: {num_bits: 8, type: int, symmetric: true, strategy: channel}
+          input_activations: {num_bits: 8, type: int, symmetric: true, dynamic: true, strategy: token}
           targets: [Linear]
-    SparseGPTModifier:
-      sparsity: 0.0
-      block_size: 128
-      sequential_update: false
-      quantize: true
-      targets: ['re:model.layers.\d+$']

tokenizer.json

@@ -134,6 +134,7 @@
     "end_of_word_suffix": null,
     "fuse_unk": true,
     "byte_fallback": true,
+    "ignore_merges": false,
     "vocab": {
       "<unk>": 0,
       "<s>": 1,

tokenizer_config.json

@@ -28,6 +28,7 @@
     }
   },
   "bos_token": "<s>",
+  "chat_template": "{% for message in messages %}\n{% if message['role'] == 'user' %}\n{{ '<|user|>\n' + message['content'] + eos_token }}\n{% elif message['role'] == 'system' %}\n{{ '<|system|>\n' + message['content'] + eos_token }}\n{% elif message['role'] == 'assistant' %}\n{{ '<|assistant|>\n'  + message['content'] + eos_token }}\n{% endif %}\n{% if loop.last and add_generation_prompt %}\n{{ '<|assistant|>' }}\n{% endif %}\n{% endfor %}",
   "clean_up_tokenization_spaces": false,
   "eos_token": "</s>",
   "legacy": false,