Training in progress, epoch 1

config.json

@@ -1,34 +1,35 @@
-{
-  "_name_or_path": "microsoft/git-base",
-  "architectures": [
-    "GitForCausalLM"
-  ],
-  "attention_probs_dropout_prob": 0.1,
-  "bos_token_id": 101,
-  "classifier_dropout": null,
-  "eos_token_id": 102,
-  "hidden_act": "gelu",
-  "hidden_dropout_prob": 0.1,
-  "hidden_size": 768,
-  "initializer_range": 0.02,
-  "intermediate_size": 3072,
-  "layer_norm_eps": 1e-12,
-  "max_position_embeddings": 1024,
-  "model_type": "git",
-  "num_attention_heads": 12,
-  "num_hidden_layers": 6,
-  "num_image_with_embedding": null,
-  "pad_token_id": 0,
-  "position_embedding_type": "absolute",
-  "tie_word_embeddings": false,
-  "torch_dtype": "float32",
-  "transformers_version": "4.45.2",
-  "use_cache": true,
-  "vision_config": {
-    "dropout": 0.0,
-    "initializer_factor": 1.0,
-    "model_type": "git_vision_model",
-    "projection_dim": 512
-  },
-  "vocab_size": 30522
-}
+{
+  "_name_or_path": "microsoft/git-base",
+  "architectures": [
+    "GitForCausalLM"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "bos_token_id": 101,
+  "classifier_dropout": null,
+  "eos_token_id": 102,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 1024,
+  "model_type": "git",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 6,
+  "num_image_with_embedding": null,
+  "pad_token_id": 0,
+  "position_embedding_type": "absolute",
+  "tie_word_embeddings": false,
+  "torch_dtype": "float32",
+  "transformers_version": "4.47.1",
+  "use_cache": true,
+  "vision_config": {
+    "_attn_implementation_autoset": true,
+    "dropout": 0.0,
+    "initializer_factor": 1.0,
+    "model_type": "git_vision_model",
+    "projection_dim": 512
+  },
+  "vocab_size": 30522
+}

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e6fb48ec37ebddf2f8f502a6be03da96cfa3b194de5ef4658c6383ff1ab786f3
+oid sha256:fa34cf5940b66cb4968c73c48b1e6eedc0d7d8eccd3ab4f83ce24cee92197808
 size 706516040

runs/Dec29_23-39-48_4d6eb9645fff/events.out.tfevents.1735515589.4d6eb9645fff.11139.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f6960d3be546436259e61ec500e17c177f697c688a0e77b21d873ec52228b323
+size 5525

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a97e8445887d40281b4cf415419625111fbe9c5468ab9b2d363c574b92475cf5
-size 5240
+oid sha256:afd10b3dffa5a4acc3d4a8308ebaa7d27e2b666918d14ad91308cae5444280ec
+size 5368