Upload Qwen3ForCausalLM

config.json

@@ -5,7 +5,7 @@
   "attention_bias": false,
   "attention_dropout": 0.0,
   "bos_token_id": 151643,
-  "dtype": "float32",
+  "dtype": "bfloat16",
   "eos_token_id": 151645,
   "head_dim": 128,
   "hidden_act": "silu",
@@ -56,12 +56,15 @@
   "num_attention_heads": 32,
   "num_hidden_layers": 36,
   "num_key_value_heads": 8,
+  "pad_token_id": null,
   "rms_norm_eps": 1e-06,
-  "rope_scaling": null,
-  "rope_theta": 5000000,
+  "rope_parameters": {
+    "rope_theta": 5000000,
+    "rope_type": "default"
+  },
   "sliding_window": null,
   "tie_word_embeddings": true,
-  "transformers_version": "4.56.1",
+  "transformers_version": "5.3.0",
   "use_cache": true,
   "use_sliding_window": false,
   "vocab_size": 172149

generation_config.json

@@ -9,5 +9,5 @@
   "temperature": 0.7,
   "top_k": 20,
   "top_p": 0.8,
-  "transformers_version": "4.56.1"
+  "transformers_version": "5.3.0"
 }

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d5f148367f0927ff1661a29b60ca77bb7b5d5984878625edc8b266f467ac8458
+size 8148472656