Upload folder using huggingface_hub

config.json

@@ -103,9 +103,11 @@
           "num_bits": 8,
           "observer": "mse",
           "observer_kwargs": {},
+          "scale_dtype": null,
           "strategy": "group",
           "symmetric": true,
-          "type": "int"
+          "type": "int",
+          "zp_dtype": null
         }
       }
     },
@@ -258,6 +260,9 @@
       "model.layers.46.mlp.shared_experts.gate_proj",
       "model.layers.46.mlp.shared_experts.up_proj",
       "model.layers.46.mlp.shared_experts.down_proj",
+      "model.layers.47.mlp.shared_experts.gate_proj",
+      "model.layers.47.mlp.shared_experts.up_proj",
+      "model.layers.47.mlp.shared_experts.down_proj",
       "lm_head"
     ],
     "kv_cache_scheme": null,
@@ -265,7 +270,7 @@
     "quantization_status": "compressed",
     "sparsity_config": {},
     "transform_config": {},
-    "version": "0.13.0"
+    "version": "0.13.1.a20260219"
   },
   "rms_norm_eps": 1e-05,
   "rope_interleave": true,
@@ -278,7 +283,7 @@
   "tie_word_embeddings": false,
   "topk_group": 1,
   "topk_method": "noaux_tc",
-  "transformers_version": "5.0.0.dev0",
+  "transformers_version": "5.3.0.dev0",
   "use_cache": true,
   "v_head_dim": 256,
   "vocab_size": 154880

generation_config.json

@@ -7,5 +7,5 @@
   ],
   "pad_token_id": 154820,
   "temperature": 1.0,
-  "transformers_version": "5.0.0.dev0"
+  "transformers_version": "5.3.0.dev0"
 }

model-00001-of-00007.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:861f47e51bc629830d1735a033d7a67168cb4ca93789e866a4306a7a37f31252
-size 5369059752
+oid sha256:8e6de34dd979aa0236eb5a46dcd4b72b786b232e84ff70719fbf3032dcbf53a6
+size 5366585992

model-00002-of-00007.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2f1295be294c1a3590eedffe56c7323a404daaeb5a913a043869e2e818bcd668
-size 5367617216
+oid sha256:ab495f8d2028bf45a8d0982352282342f4da9fb3534de6b91bf35fa4152bca4a
+size 5368942680

model-00003-of-00007.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:680ce17f22283e8c86728fa0420544671daf22c6fc3d9c56399552f425cb98be
-size 5367879648
+oid sha256:bb9f921e767856daa24c3bd3e58b24ff4f2d6398cc5aea720a04ff9cacee0340
+size 5368943280

model-00004-of-00007.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2313c101e382208b4387262bd82db7fd5458dfb8016954412b9af187fe37ecd8
-size 5367617216
+oid sha256:f19279805405a2ae516bd35e5e5e526be00c24a4f33ba75f3f4fa595f5702030
+size 5368942688

model-00005-of-00007.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:840456628de3f244f218e44ec72f3a7344a443776f1d04e8a798d15855e6ad56
-size 5367879464
+oid sha256:6f29d1d20fc48f0b0e7178fd6fe54a4502daf77e81e3ad266e07deed81b6a710
+size 5368942680

model-00006-of-00007.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4a3e8affff5a8631b73a05564e240303b6227cef2f30104a9d4fb94d1aa88c3b
-size 5369511400
+oid sha256:93ee0d52e6b78c6030b96cd87a32930b7d9e5e3187bcae4bc173e41ec2bfb615
+size 5366592656

model-00007-of-00007.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1a3a28eba6e155024ed66e5163c21824922d9bc060d75020148a338a20dc4d16
-size 695839176
+oid sha256:4e8c0b336575cb6e11c887a749097aeafe1926fd7b3085f4d09ef989146f1f3b
+size 2666083272

recipe.yaml

@@ -21,20 +21,26 @@ default_stage:
           output_activations: null
           format: null
       targets: [Linear]
-      ignore: [lm_head, model.embed_tokens, 're:.*input_layernorm$', 're:.*post_attention_layernorm$',
-        model.norm, 're:.*shared_experts.*', 're:model[.]layers[.]0[.].*', 're:.*mlp[.]gate$']
+      ignore: [lm_head, model.embed_tokens, 're:.*shared_experts.*', 're:model[.]layers[.]0[.].*',
+        're:.*mlp[.]gate$']
+      bypass_divisibility_checks: false
       mappings:
-      - smooth_layer: re:.*input_layernorm$
-        balance_layers: ['re:.*(q_a_proj|q_proj)$', 're:.*kv_a_proj_with_mqa$']
-      - smooth_layer: re:.*q_a_layernorm$
-        balance_layers: ['re:.*q_b_proj$']
-      - smooth_layer: re:.*kv_a_layernorm$
-        balance_layers: ['re:.*kv_b_proj$']
-      - smooth_layer: re:.*post_attention_layernorm$
-        balance_layers: ['re:.*gate_proj$', 're:.*up_proj$', 're:.*gate_up_proj$']
-      - smooth_layer: re:.*up_proj$
-        balance_layers: ['re:.*down_proj$']
-      - smooth_layer: re:.*gate_up_proj$
-        balance_layers: ['re:.*down_proj$']
+      - smooth_layer: re:.*layers[.]((?:[1-9]|[1-3][0-9]|4[0-6]))[.].*input_layernorm$
+        balance_layers: ['re:.*layers[.]((?:[1-9]|[1-3][0-9]|4[0-6]))[.].*q_a_proj$', 're:.*layers[.]((?:[1-9]|[1-3][0-9]|4[0-6]))[.].*kv_a_proj_with_mqa$']
+        activation_hook_target: null
+      - smooth_layer: re:.*layers[.]((?:[1-9]|[1-3][0-9]|4[0-6]))[.].*q_a_layernorm$
+        balance_layers: ['re:.*layers[.]((?:[1-9]|[1-3][0-9]|4[0-6]))[.].*q_b_proj$']
+        activation_hook_target: null
+      - smooth_layer: re:.*layers[.]((?:[1-9]|[1-3][0-9]|4[0-6]))[.].*kv_a_layernorm$
+        balance_layers: ['re:.*layers[.]((?:[1-9]|[1-3][0-9]|4[0-6]))[.].*kv_b_proj$']
+        activation_hook_target: null
+      - smooth_layer: re:.*layers[.]((?:[1-9]|[1-3][0-9]|4[0-6]))[.].*post_attention_layernorm$
+        balance_layers: ['re:.*layers[.]((?:[1-9]|[1-3][0-9]|4[0-6]))[.].*gate_proj$', 're:.*layers[.]((?:[1-9]|[1-3][0-9]|4[0-6]))[.].*up_proj$',
+          're:.*layers[.]((?:[1-9]|[1-3][0-9]|4[0-6]))[.].*mlp[.]gate$']
+        activation_hook_target: null
+      - smooth_layer: re:.*layers[.]((?:[1-9]|[1-3][0-9]|4[0-6]))[.].*mlp[.]experts.*up_proj$
+        balance_layers: ['re:.*layers[.]((?:[1-9]|[1-3][0-9]|4[0-6]))[.].*mlp[.]experts.*down_proj$']
+        activation_hook_target: null
+      offload_device: !!python/object/apply:torch.device [cuda]
       duo_scaling: true
       n_grid: 20

tokenizer.json

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3195aaa8cc306d5112246924f5ad05905f6bd28b02c29a955e2f870e4f8ff3fc
-size 20217540
+oid sha256:353a3ff93723c098ac93fb593d4d1b93faede916fe502b005cd503871023e012
+size 20217541

tokenizer_config.json

@@ -3,29 +3,8 @@
   "clean_up_tokenization_spaces": false,
   "do_lower_case": false,
   "eos_token": "<|endoftext|>",
-  "extra_special_tokens": [
-    "<|endoftext|>",
-    "[MASK]",
-    "[gMASK]",
-    "[sMASK]",
-    "<sop>",
-    "<eop>",
-    "<|system|>",
-    "<|user|>",
-    "<|assistant|>",
-    "<|observation|>",
-    "<|begin_of_image|>",
-    "<|end_of_image|>",
-    "<|begin_of_video|>",
-    "<|end_of_video|>",
-    "<|begin_of_audio|>",
-    "<|end_of_audio|>",
-    "<|begin_of_transcription|>",
-    "<|end_of_transcription|>"
-  ],
   "is_local": false,
   "model_max_length": 128000,
-  "model_specific_special_tokens": {},
   "pad_token": "<|endoftext|>",
   "padding_side": "left",
   "remove_space": false,