Training in progress, step 500

config.json

@@ -3,7 +3,6 @@
     "VisionEncoderDecoderModel"
   ],
   "decoder": {
-    "_attn_implementation_autoset": true,
     "_name_or_path": "gpt2",
     "activation_function": "gelu_new",
     "add_cross_attention": true,
@@ -44,7 +43,6 @@
   "decoder_start_token_id": 50256,
   "early_stopping": null,
   "encoder": {
-    "_attn_implementation_autoset": true,
     "_name_or_path": "facebook/timesformer-base-finetuned-k600",
     "architectures": [
       "TimesformerForVideoClassification"
@@ -1272,6 +1270,7 @@
     "qkv_bias": true,
     "torch_dtype": "float32"
   },
+  "eos_token_id": 50256,
   "is_encoder_decoder": true,
   "max_length": null,
   "model_type": "vision-encoder-decoder",

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d3e47c4e369484e88b24290baad37862866f250988bddce4151d94df06d8c553
+oid sha256:2364c59a5ece25dcce95c5af4cf3396fd5ddb5fc34954d439e3ac96c1d7e945e
 size 1096319528

runs/Apr11_23-11-22_f87f15f6f192/events.out.tfevents.1744413083.f87f15f6f192.31.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:42d1f76a0507f42154f66d31d15e462a6a9ed8229d9bac22d09d295a9c1917bf
+size 43505

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:062e5eade887ffde47fa7a2ecfb345b935a5dcebd3948ad6f29d72144a307f96
+oid sha256:d513cd4278464418cae5f56ee8fa9a7d62a31a64be52a3627ce38011004c9f3e
 size 5496