Upload GotOcr2ForConditionalGeneration

config.json

@@ -2,29 +2,25 @@
   "architectures": [
     "GotOcr2ForConditionalGeneration"
   ],
-  "attention_dropout": 0.0,
-  "hidden_act": "silu",
-  "hidden_size": 1024,
-  "image_token_id": 151859,
-  "initializer_range": 0.02,
-  "intermediate_size": 2816,
-  "max_position_embeddings": 32768,
-  "max_window_layers": 21,
+  "ignore_index": -100,
+  "image_seq_length": 576,
+  "image_token_index": 151859,
   "model_type": "got_ocr2",
-  "num_attention_heads": 16,
-  "num_hidden_layers": 24,
-  "num_key_value_heads": 16,
-  "rms_norm_eps": 1e-06,
-  "rope_scaling": null,
-  "rope_theta": 1000000.0,
-  "sliding_window": 32768,
-  "tie_word_embeddings": true,
+  "text_config": {
+    "hidden_size": 1024,
+    "intermediate_size": 2816,
+    "max_window_layers": 21,
+    "model_type": "qwen2",
+    "num_attention_heads": 16,
+    "num_hidden_layers": 24,
+    "num_key_value_heads": 16,
+    "rope_theta": 1000000.0,
+    "tie_word_embeddings": true,
+    "vocab_size": 151860
+  },
   "torch_dtype": "bfloat16",
   "transformers_version": "4.49.0.dev0",
-  "use_cache": true,
-  "use_sliding_window": false,
   "vision_config": {
     "model_type": "got_ocr2_vision_model"
-  },
-  "vocab_size": 151860
+  }
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5b140b2b64af7a3505c4fe7406d5f9fbd436213ca0be1b9460e16c0d9c47af97
-size 1121113328
+oid sha256:6175ac7868a4e75735f5d59f78c465081ad3427eb4f312d072a0f1d16b333ba4
+size 1121114488