Upload Guardrail Classifier

config.json

@@ -0,0 +1,34 @@
+{
+  "activation": "gelu",
+  "architectures": [
+    "DistilBertForSequenceClassification"
+  ],
+  "attention_dropout": 0.1,
+  "dim": 768,
+  "dropout": 0.1,
+  "dtype": "float32",
+  "hidden_dim": 3072,
+  "id2label": {
+    "0": "TASK_1",
+    "1": "TASK_2",
+    "2": "TASK_3"
+  },
+  "initializer_range": 0.02,
+  "label2id": {
+    "TASK_1": 0,
+    "TASK_2": 1,
+    "TASK_3": 2
+  },
+  "max_position_embeddings": 512,
+  "model_type": "distilbert",
+  "n_heads": 12,
+  "n_layers": 6,
+  "pad_token_id": 0,
+  "problem_type": "single_label_classification",
+  "qa_dropout": 0.1,
+  "seq_classif_dropout": 0.2,
+  "sinusoidal_pos_embds": false,
+  "tie_weights_": true,
+  "transformers_version": "4.57.1",
+  "vocab_size": 30522
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b0f13e9a76ad4b33f9d4d883fbc29755f3a50bb7b06c0367b37ee6a9afa763f3
+size 267835644

optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:53444c60af2cda9fbca728e9cc875ffe4786bc85d757ef55a3ccb5594a1b1ef0
+size 535735371

rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5902280ce690673c5322e3954823647484a8b352a7ff28074852f0b72bbea4d5
+size 14645

scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:236f57790f8647a516281c7e83d121658ecb52e1a557554b78e18f5daebb0d7d
+size 1465

special_tokens_map.json

@@ -0,0 +1,7 @@
+{
+  "cls_token": "[CLS]",
+  "mask_token": "[MASK]",
+  "pad_token": "[PAD]",
+  "sep_token": "[SEP]",
+  "unk_token": "[UNK]"
+}

tokenizer_config.json

@@ -0,0 +1,56 @@
+{
+  "added_tokens_decoder": {
+    "0": {
+      "content": "[PAD]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "100": {
+      "content": "[UNK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "101": {
+      "content": "[CLS]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "102": {
+      "content": "[SEP]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "103": {
+      "content": "[MASK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "clean_up_tokenization_spaces": false,
+  "cls_token": "[CLS]",
+  "do_lower_case": true,
+  "extra_special_tokens": {},
+  "mask_token": "[MASK]",
+  "model_max_length": 512,
+  "pad_token": "[PAD]",
+  "sep_token": "[SEP]",
+  "strip_accents": null,
+  "tokenize_chinese_chars": true,
+  "tokenizer_class": "DistilBertTokenizer",
+  "unk_token": "[UNK]"
+}

trainer_state.json

@@ -0,0 +1,161 @@
+{
+  "best_global_step": 570,
+  "best_metric": 0.0015665935352444649,
+  "best_model_checkpoint": "./guardrail_model\\checkpoint-570",
+  "epoch": 10.0,
+  "eval_steps": 500,
+  "global_step": 950,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.9940476190476191,
+      "eval_f1": 0.9941583470995236,
+      "eval_loss": 0.038976069539785385,
+      "eval_precision": 0.9956709956709956,
+      "eval_recall": 0.9927536231884058,
+      "eval_runtime": 0.2694,
+      "eval_samples_per_second": 623.529,
+      "eval_steps_per_second": 77.941,
+      "step": 95
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.9940476190476191,
+      "eval_f1": 0.9942082650905553,
+      "eval_loss": 0.018449116498231888,
+      "eval_precision": 0.9929078014184397,
+      "eval_recall": 0.9956140350877193,
+      "eval_runtime": 0.5024,
+      "eval_samples_per_second": 334.383,
+      "eval_steps_per_second": 41.798,
+      "step": 190
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 1.0,
+      "eval_f1": 1.0,
+      "eval_loss": 0.005749495234340429,
+      "eval_precision": 1.0,
+      "eval_recall": 1.0,
+      "eval_runtime": 0.4167,
+      "eval_samples_per_second": 403.171,
+      "eval_steps_per_second": 50.396,
+      "step": 285
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.9880952380952381,
+      "eval_f1": 0.9884633569739952,
+      "eval_loss": 0.02493114396929741,
+      "eval_precision": 0.9861111111111112,
+      "eval_recall": 0.9912280701754387,
+      "eval_runtime": 0.3582,
+      "eval_samples_per_second": 468.954,
+      "eval_steps_per_second": 58.619,
+      "step": 380
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.9940476190476191,
+      "eval_f1": 0.9927527669463153,
+      "eval_loss": 0.009493241086602211,
+      "eval_precision": 0.9929078014184397,
+      "eval_recall": 0.9927536231884058,
+      "eval_runtime": 0.256,
+      "eval_samples_per_second": 656.141,
+      "eval_steps_per_second": 82.018,
+      "step": 475
+    },
+    {
+      "epoch": 5.2631578947368425,
+      "grad_norm": 0.01788582280278206,
+      "learning_rate": 9.494736842105265e-06,
+      "loss": 0.0837,
+      "step": 500
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 1.0,
+      "eval_f1": 1.0,
+      "eval_loss": 0.0015665935352444649,
+      "eval_precision": 1.0,
+      "eval_recall": 1.0,
+      "eval_runtime": 0.2619,
+      "eval_samples_per_second": 641.401,
+      "eval_steps_per_second": 80.175,
+      "step": 570
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 1.0,
+      "eval_f1": 1.0,
+      "eval_loss": 0.003049974562600255,
+      "eval_precision": 1.0,
+      "eval_recall": 1.0,
+      "eval_runtime": 0.2578,
+      "eval_samples_per_second": 651.735,
+      "eval_steps_per_second": 81.467,
+      "step": 665
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 1.0,
+      "eval_f1": 1.0,
+      "eval_loss": 0.004172018729150295,
+      "eval_precision": 1.0,
+      "eval_recall": 1.0,
+      "eval_runtime": 0.2574,
+      "eval_samples_per_second": 652.703,
+      "eval_steps_per_second": 81.588,
+      "step": 760
+    },
+    {
+      "epoch": 9.0,
+      "eval_accuracy": 1.0,
+      "eval_f1": 1.0,
+      "eval_loss": 0.004257123451679945,
+      "eval_precision": 1.0,
+      "eval_recall": 1.0,
+      "eval_runtime": 0.2587,
+      "eval_samples_per_second": 649.434,
+      "eval_steps_per_second": 81.179,
+      "step": 855
+    },
+    {
+      "epoch": 10.0,
+      "eval_accuracy": 1.0,
+      "eval_f1": 1.0,
+      "eval_loss": 0.004139341879636049,
+      "eval_precision": 1.0,
+      "eval_recall": 1.0,
+      "eval_runtime": 0.2544,
+      "eval_samples_per_second": 660.397,
+      "eval_steps_per_second": 82.55,
+      "step": 950
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 950,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 10,
+  "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": 101364981453558.0,
+  "train_batch_size": 16,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:def71aa6eee12e899711aaa59222fab129e848408b3e3d4cf3852760255c3b9f
+size 5841