Upload 12 files

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+tokenizer.json filter=lfs diff=lfs merge=lfs -text

added_tokens.json

@@ -0,0 +1,3 @@
+{
+  "<image_soft_token>": 262144
+}

all_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 3.0,
+    "eval_accuracy": 0.8978077571669477,
+    "eval_loss": 0.2924116253852844,
+    "eval_runtime": 7.2749,
+    "eval_samples": 2965,
+    "eval_samples_per_second": 407.564,
+    "eval_steps_per_second": 50.997,
+    "num_input_tokens_seen": 4552704,
+    "total_flos": 2773636063690752.0,
+    "train_loss": 0.30860748496937207,
+    "train_runtime": 450.538,
+    "train_samples": 11856,
+    "train_samples_per_second": 78.946,
+    "train_steps_per_second": 9.868
+}

config.json

@@ -0,0 +1,79 @@
+{
+  "_sliding_window_pattern": 6,
+  "architectures": [
+    "Gemma3TextForSequenceClassification"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "attn_logit_softcapping": null,
+  "bos_token_id": 2,
+  "dtype": "float32",
+  "eos_token_id": 1,
+  "final_logit_softcapping": null,
+  "finetuning_task": "text-classification",
+  "head_dim": 256,
+  "hidden_activation": "gelu_pytorch_tanh",
+  "hidden_size": 768,
+  "id2label": {
+    "0": "False",
+    "1": "True"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 1152,
+  "label2id": {
+    "False": 0,
+    "True": 1
+  },
+  "layer_types": [
+    "sliding_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "full_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "full_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "full_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "full_attention"
+  ],
+  "max_position_embeddings": 2048,
+  "model_type": "gemma3_text",
+  "num_attention_heads": 3,
+  "num_hidden_layers": 24,
+  "num_key_value_heads": 1,
+  "pad_token_id": 0,
+  "problem_type": "single_label_classification",
+  "query_pre_attn_scalar": 256,
+  "rms_norm_eps": 1e-06,
+  "rope_local_base_freq": 10000.0,
+  "rope_parameters": {
+    "full_attention": {
+      "rope_theta": 1000000.0,
+      "rope_type": "default"
+    },
+    "sliding_attention": {
+      "rope_theta": 10000.0,
+      "rope_type": "default"
+    }
+  },
+  "rope_theta": 1000000.0,
+  "sliding_window": 257,
+  "transformers_version": "5.0.0.dev0",
+  "use_bidirectional_attention": true,
+  "use_cache": false,
+  "vocab_size": 262144
+}

eval_results.json

@@ -0,0 +1,10 @@
+{
+    "epoch": 3.0,
+    "eval_accuracy": 0.8978077571669477,
+    "eval_loss": 0.2924116253852844,
+    "eval_runtime": 7.2749,
+    "eval_samples": 2965,
+    "eval_samples_per_second": 407.564,
+    "eval_steps_per_second": 50.997,
+    "num_input_tokens_seen": 4552704
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7ebcc7e2c88fed34161f1dfeefc6e633acc0b458aee3b326ccf349ae8a1c144d
+size 1211494184

special_tokens_map.json

@@ -0,0 +1,33 @@
+{
+  "boi_token": "<start_of_image>",
+  "bos_token": {
+    "content": "<bos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eoi_token": "<end_of_image>",
+  "eos_token": {
+    "content": "<eos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "image_token": "<image_soft_token>",
+  "pad_token": {
+    "content": "<pad>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6852f8d561078cc0cebe70ca03c5bfdd0d60a45f9d2e0e1e4cc05b68e9ec329e
+size 33385008

tokenizer.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1299c11d7cf632ef3b4e11937501358ada021bbdf7c47638d13c0ee982f2e79c
+size 4689074

train_results.json

@@ -0,0 +1,10 @@
+{
+    "epoch": 3.0,
+    "num_input_tokens_seen": 4552704,
+    "total_flos": 2773636063690752.0,
+    "train_loss": 0.30860748496937207,
+    "train_runtime": 450.538,
+    "train_samples": 11856,
+    "train_samples_per_second": 78.946,
+    "train_steps_per_second": 9.868
+}

trainer_state.json

@@ -0,0 +1,154 @@
+{
+  "best_global_step": 2964,
+  "best_metric": 0.2924116253852844,
+  "best_model_checkpoint": "embeddinggemma-300m-pii-detector/checkpoint-2964",
+  "epoch": 3.0,
+  "eval_steps": 500,
+  "global_step": 4446,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.33738191632928477,
+      "grad_norm": 17.098201751708984,
+      "learning_rate": 4.438821412505623e-05,
+      "loss": 0.5645,
+      "num_input_tokens_seen": 512000,
+      "step": 500,
+      "train_runtime": 47.9743,
+      "train_tokens_per_second": 10672.374
+    },
+    {
+      "epoch": 0.6747638326585695,
+      "grad_norm": 1.4570860862731934,
+      "learning_rate": 3.876518218623482e-05,
+      "loss": 0.3907,
+      "num_input_tokens_seen": 1024000,
+      "step": 1000,
+      "train_runtime": 94.2729,
+      "train_tokens_per_second": 10862.077
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.8627318718381113,
+      "eval_loss": 0.30147919058799744,
+      "eval_runtime": 7.391,
+      "eval_samples_per_second": 401.161,
+      "eval_steps_per_second": 50.196,
+      "num_input_tokens_seen": 1517568,
+      "step": 1482
+    },
+    {
+      "epoch": 1.0121457489878543,
+      "grad_norm": 14.889694213867188,
+      "learning_rate": 3.3142150247413403e-05,
+      "loss": 0.3335,
+      "num_input_tokens_seen": 1536000,
+      "step": 1500,
+      "train_runtime": 151.9538,
+      "train_tokens_per_second": 10108.336
+    },
+    {
+      "epoch": 1.349527665317139,
+      "grad_norm": 0.4145456552505493,
+      "learning_rate": 2.7519118308591997e-05,
+      "loss": 0.3069,
+      "num_input_tokens_seen": 2048000,
+      "step": 2000,
+      "train_runtime": 198.9267,
+      "train_tokens_per_second": 10295.252
+    },
+    {
+      "epoch": 1.686909581646424,
+      "grad_norm": 49.79447555541992,
+      "learning_rate": 2.1896086369770583e-05,
+      "loss": 0.2847,
+      "num_input_tokens_seen": 2560000,
+      "step": 2500,
+      "train_runtime": 245.6972,
+      "train_tokens_per_second": 10419.327
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.8978077571669477,
+      "eval_loss": 0.2924116253852844,
+      "eval_runtime": 7.3471,
+      "eval_samples_per_second": 403.563,
+      "eval_steps_per_second": 50.496,
+      "num_input_tokens_seen": 3035136,
+      "step": 2964
+    },
+    {
+      "epoch": 2.0242914979757085,
+      "grad_norm": 0.2802899181842804,
+      "learning_rate": 1.627305443094917e-05,
+      "loss": 0.2542,
+      "num_input_tokens_seen": 3072000,
+      "step": 3000,
+      "train_runtime": 303.072,
+      "train_tokens_per_second": 10136.207
+    },
+    {
+      "epoch": 2.361673414304993,
+      "grad_norm": 29.810983657836914,
+      "learning_rate": 1.0650022492127757e-05,
+      "loss": 0.2176,
+      "num_input_tokens_seen": 3584000,
+      "step": 3500,
+      "train_runtime": 350.1674,
+      "train_tokens_per_second": 10235.104
+    },
+    {
+      "epoch": 2.699055330634278,
+      "grad_norm": 14.638989448547363,
+      "learning_rate": 5.026990553306343e-06,
+      "loss": 0.2064,
+      "num_input_tokens_seen": 4096000,
+      "step": 4000,
+      "train_runtime": 397.7303,
+      "train_tokens_per_second": 10298.436
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.8930860033726813,
+      "eval_loss": 0.30713844299316406,
+      "eval_runtime": 7.2761,
+      "eval_samples_per_second": 407.498,
+      "eval_steps_per_second": 50.989,
+      "num_input_tokens_seen": 4552704,
+      "step": 4446
+    },
+    {
+      "epoch": 3.0,
+      "num_input_tokens_seen": 4552704,
+      "step": 4446,
+      "total_flos": 2773636063690752.0,
+      "train_loss": 0.30860748496937207,
+      "train_runtime": 450.538,
+      "train_samples_per_second": 78.946,
+      "train_steps_per_second": 9.868
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 4446,
+  "num_input_tokens_seen": 4552704,
+  "num_train_epochs": 3,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 2773636063690752.0,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8ac0e0ebd3184ed9e9684938804ccd3f3ef2b199540b7c3e8e7d524600e73af4
+size 5201