(Trained with Unsloth)

added_tokens.json

@@ -1,7 +1,10 @@
 {
+  "</think>": 151668,
   "</tool_call>": 151658,
+  "</tool_response>": 151666,
+  "<think>": 151667,
   "<tool_call>": 151657,
-  "<|PAD_TOKEN|>": 151665,
+  "<tool_response>": 151665,
   "<|box_end|>": 151649,
   "<|box_start|>": 151648,
   "<|endoftext|>": 151643,

config.json

@@ -1,31 +1,32 @@
 {
   "architectures": [
-    "Qwen2ForCausalLM"
+    "Qwen3ForCausalLM"
   ],
+  "attention_bias": false,
   "attention_dropout": 0.0,
-  "bos_token_id": 151643,
   "eos_token_id": 151643,
+  "head_dim": 128,
   "hidden_act": "silu",
-  "hidden_size": 3584,
+  "hidden_size": 2560,
   "initializer_range": 0.02,
-  "intermediate_size": 18944,
+  "intermediate_size": 9728,
   "max_position_embeddings": 32768,
-  "max_window_layers": 28,
-  "model_type": "qwen2",
-  "num_attention_heads": 28,
-  "num_hidden_layers": 28,
-  "num_key_value_heads": 4,
-  "pad_token_id": 151665,
+  "max_window_layers": 36,
+  "model_type": "qwen3",
+  "num_attention_heads": 32,
+  "num_hidden_layers": 36,
+  "num_key_value_heads": 8,
+  "pad_token_id": 151654,
   "rms_norm_eps": 1e-06,
   "rope_scaling": null,
-  "rope_theta": 1000000.0,
+  "rope_theta": 1000000,
   "sliding_window": null,
-  "tie_word_embeddings": false,
+  "tie_word_embeddings": true,
   "torch_dtype": "bfloat16",
   "transformers_version": "4.52.4",
   "unsloth_fixed": true,
   "unsloth_version": "2025.6.2",
   "use_cache": true,
   "use_sliding_window": false,
-  "vocab_size": 152064
+  "vocab_size": 151936
 }

generation_config.json

@@ -3,6 +3,6 @@
   "eos_token_id": 151643,
   "max_length": 32768,
   "max_new_tokens": 2048,
-  "pad_token_id": 151665,
+  "pad_token_id": 151654,
   "transformers_version": "4.52.4"
 }

special_tokens_map.json

@@ -22,7 +22,7 @@
     "single_word": false
   },
   "pad_token": {
-    "content": "<|PAD_TOKEN|>",
+    "content": "<|vision_pad|>",
     "lstrip": false,
     "normalized": false,
     "rstrip": false,

tokenizer.json

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fab42efe8d17406525a9154b728cf9e957629a8ed7ce997770efdd71128c6a1a
-size 11422086
+oid sha256:aeb13307a71acd8fe81861d94ad54ab689df773318809eed3cbe794b4492dae4
+size 11422654

tokenizer_config.json

@@ -179,12 +179,36 @@
       "special": false
     },
     "151665": {
-      "content": "<|PAD_TOKEN|>",
+      "content": "<tool_response>",
       "lstrip": false,
       "normalized": false,
       "rstrip": false,
       "single_word": false,
-      "special": true
+      "special": false
+    },
+    "151666": {
+      "content": "</tool_response>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151667": {
+      "content": "<think>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151668": {
+      "content": "</think>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
     }
   },
   "additional_special_tokens": [
@@ -208,7 +232,7 @@
   "errors": "replace",
   "extra_special_tokens": {},
   "model_max_length": 32768,
-  "pad_token": "<|PAD_TOKEN|>",
+  "pad_token": "<|vision_pad|>",
   "padding_side": "right",
   "split_special_tokens": false,
   "tokenizer_class": "Qwen2Tokenizer",