Training in progress, step 2

.ipynb_checkpoints/config-checkpoint.json

@@ -0,0 +1,40 @@
+{
+  "_name_or_path": "google/paligemma-3b-pt-224",
+  "architectures": [
+    "PaliGemmaForConditionalGeneration"
+  ],
+  "bos_token_id": 2,
+  "eos_token_id": 1,
+  "hidden_size": 2048,
+  "ignore_index": -100,
+  "image_token_index": 257152,
+  "model_type": "paligemma",
+  "pad_token_id": 0,
+  "projection_dim": 2048,
+  "text_config": {
+    "hidden_size": 2048,
+    "intermediate_size": 16384,
+    "model_type": "gemma",
+    "num_attention_heads": 8,
+    "num_hidden_layers": 18,
+    "num_image_tokens": 256,
+    "num_key_value_heads": 1,
+    "torch_dtype": "float32",
+    "vocab_size": 257216
+  },
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.42.0.dev0",
+  "vision_config": {
+    "hidden_size": 1152,
+    "intermediate_size": 4304,
+    "model_type": "siglip_vision_model",
+    "num_attention_heads": 16,
+    "num_hidden_layers": 27,
+    "num_image_tokens": 256,
+    "patch_size": 14,
+    "projection_dim": 2048,
+    "projector_hidden_act": "gelu_fast",
+    "vision_use_head": false
+  },
+  "vocab_size": 257216
+}

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "google/paligemma-3b-pt-224",
+  "_name_or_path": "google/paligemma-3b-mix-224",
   "architectures": [
     "PaliGemmaForConditionalGeneration"
   ],

runs/Jun20_09-25-11_aid-iad-bm-gpu4-8-ad1-002/events.out.tfevents.1718875520.aid-iad-bm-gpu4-8-ad1-002.863615.6

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:690734e198e0691d1a8614aa100eb0940494ac1881fe4a214c9be0cf89734175
+size 5213

runs/Jun20_09-27-35_aid-iad-bm-gpu4-8-ad1-002/events.out.tfevents.1718875668.aid-iad-bm-gpu4-8-ad1-002.877930.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ab6daaab36c37962bd468716cdcb41c9648fd15f9c35cc3b09a9c863f9d0f843
+size 5213

runs/Jun20_09-35-40_aid-iad-bm-gpu4-8-ad1-002/events.out.tfevents.1718876149.aid-iad-bm-gpu4-8-ad1-002.878270.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5f8d5257ea6579750329c5600a94a606929564a014a8f5c5ac06bfad32db78e4
+size 5213

runs/Jun20_09-39-47_aid-iad-bm-gpu4-8-ad1-002/events.out.tfevents.1718876393.aid-iad-bm-gpu4-8-ad1-002.878270.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bf95d5be4ab208d3db74e97ac3fd3986734d1046cab4066bd0eab12a7c07c5c3
+size 5213

runs/Jun20_09-42-13_aid-iad-bm-gpu4-8-ad1-002/events.out.tfevents.1718876537.aid-iad-bm-gpu4-8-ad1-002.878270.2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1742570ae70bc9accc26d31da853d5508893465083c73b632fe2192163b950f2
+size 5560

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cf639f9407b8f19f00310ab90567cc924fb3afd40e764e6da3bfe002b3de5b50
+oid sha256:2175236a3d3154fd7afadce61667c382f28bf935b5edc8a7d25173bff884020d
 size 5112