Add action tokens and update config

config.json

@@ -12,11 +12,16 @@
     "architectures": [
       "VLlama3ForCausalLM"
     ],
+    "attention_bias": false,
+    "attention_dropout": 0.0,
     "head_dim": 64,
+    "hidden_act": "silu",
     "hidden_size": 960,
+    "initializer_range": 0.02,
     "intermediate_size": 2560,
     "is_llama_config": true,
     "max_position_embeddings": 8192,
+    "mlp_bias": false,
     "model_type": "llama",
     "neftune_noise_alpha": 0.0,
     "num_attention_heads": 15,
@@ -24,7 +29,6 @@
     "num_key_value_heads": 5,
     "pad_token_id": 2,
     "perceiver_config": {
-      "_attn_implementation_autoset": false,
       "_name_or_path": "",
       "add_cross_attention": false,
       "architectures": null,
@@ -97,17 +101,20 @@
       "use_bfloat16": false
     },
     "pixel_shuffle_factor": 4,
+    "pretraining_tp": 1,
     "qk_layer_norms": false,
     "rms_norm_eps": 1e-05,
     "rope_interleaved": false,
+    "rope_scaling": null,
     "rope_theta": 100000,
-    "torch_dtype": "bfloat16",
+    "torch_dtype": "float32",
     "transformers.js_config": {
       "kv_cache_dtype": {
         "fp16": "float16",
         "q4f16": "float16"
       }
     },
+    "use_cache": true,
     "use_resampler": false,
     "vocab_size": 49536
   },
@@ -119,22 +126,30 @@
       "q4f16": "float16"
     }
   },
-  "transformers_version": "4.47.1",
+  "transformers_version": "4.52.4",
   "use_cache": false,
   "use_reentrant_checkpointing": false,
   "vision_config": {
+    "attention_dropout": 0.0,
+    "hidden_act": "gelu_pytorch_tanh",
     "hidden_size": 768,
     "image_size": 512,
+    "initializer_range": 0.02,
+    "intermediate_size": 3072,
+    "layer_norm_eps": 1e-06,
     "max_image_size": {
       "longest_edge": 512
     },
     "model_type": "smolvlm_vision",
     "num_attention_heads": 12,
+    "num_channels": 3,
+    "num_hidden_layers": 12,
     "patch_size": 16,
     "size": {
       "longest_edge": 512
     },
     "tie_word_embeddings": false,
+    "torch_dtype": "float32",
     "use_base_siglip": false
   },
   "vocab_size": 49536

generation_config.json

@@ -3,5 +3,5 @@
   "bos_token_id": 0,
   "eos_token_id": 49279,
   "pad_token_id": 2,
-  "transformers_version": "4.47.1"
+  "transformers_version": "4.52.4"
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b9bfd456c9472c0acd5719d6e514c4b859891af205ee1a736552fd3497b8b0c3
-size 2029990624
+oid sha256:f7efc851e741bca179a92bb6492b06520df4f98cc3535868619a2ba57ea9105e
+size 2031956704