Upload Gemma3ForCausalLM

config.json

@@ -1,25 +1,32 @@
 {
   "architectures": [
-    "MistralForCausalLM"
+    "Gemma3ForCausalLM"
   ],
+  "attention_bias": false,
   "attention_dropout": 0.0,
-  "bos_token_id": 1,
+  "attn_logit_softcapping": null,
+  "bos_token_id": 2,
+  "cache_implementation": "hybrid",
   "eos_token_id": 32001,
-  "head_dim": 80,
-  "hidden_act": "silu",
+  "final_logit_softcapping": null,
+  "head_dim": 256,
+  "hidden_activation": "gelu_pytorch_tanh",
   "hidden_size": 480,
   "initializer_range": 0.02,
   "intermediate_size": 1920,
   "max_position_embeddings": 8192,
-  "model_type": "mistral",
+  "model_type": "gemma3_text",
   "num_attention_heads": 6,
   "num_hidden_layers": 32,
   "num_key_value_heads": 2,
   "pad_token_id": 32003,
+  "query_pre_attn_scalar": 256,
   "rms_norm_eps": 1e-06,
-  "rope_theta": 10000.0,
+  "rope_local_base_freq": 10000.0,
+  "rope_scaling": null,
+  "rope_theta": 1000000.0,
   "sliding_window": 4096,
-  "tie_word_embeddings": true,
+  "sliding_window_pattern": 6,
   "torch_dtype": "float32",
   "transformers_version": "4.51.3",
   "use_cache": true,

generation_config.json

@@ -1,6 +1,7 @@
 {
   "_from_model_config": true,
-  "bos_token_id": 1,
+  "bos_token_id": 2,
+  "cache_implementation": "hybrid",
   "eos_token_id": 32001,
   "pad_token_id": 32003,
   "transformers_version": "4.51.3"

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:caba12d551bed5628f3926faf2ff23966de3fc986ac0faba3055aab214859abb
-size 494169296
+oid sha256:9c82fad70e3373c9427490bcee2c85daa9886c020f27903d510232b461e5cfb9
+size 667387336