Upload: model.onnx <- model.safetensors

config.json

@@ -1,91 +1,25 @@
 {
+  "model_type": "bert",
   "architectures": [
     "BertDualHeadModel"
   ],
-  "attention_bias": false,
-  "attention_dropout": 0.0,
-  "bos_token_id": 3,
-  "classifier_activation": "silu",
-  "classifier_bias": false,
-  "classifier_dropout": 0.0,
-  "classifier_pooling": "mean",
-  "cls_token_id": 50281,
-  "decoder_bias": true,
-  "deterministic_flash_attn": false,
-  "dtype": "float32",
-  "embedding_dropout": 0.0,
-  "eos_token_id": 4,
-  "global_attn_every_n_layers": 3,
-  "gradient_checkpointing": false,
-  "hidden_activation": "gelu",
-  "hidden_size": 768,
+  "onnx_file": "model.onnx",
+  "model_format": "onnx",
+  "base_model": "Itau-Unibanco/NorBERTo-base",
+  "num_labels": 2,
   "id2label": {
     "0": "NONE",
     "1": "ENTAILMENT"
   },
-  "initializer_cutoff_factor": 2.0,
-  "initializer_range": 0.02,
-  "intermediate_size": 1152,
   "label2id": {
-    "ENTAILMENT": 1,
-    "NONE": 0
-  },
-  "layer_norm_eps": 1e-05,
-  "layer_types": [
-    "full_attention",
-    "sliding_attention",
-    "sliding_attention",
-    "full_attention",
-    "sliding_attention",
-    "sliding_attention",
-    "full_attention",
-    "sliding_attention",
-    "sliding_attention",
-    "full_attention",
-    "sliding_attention",
-    "sliding_attention",
-    "full_attention",
-    "sliding_attention",
-    "sliding_attention",
-    "full_attention",
-    "sliding_attention",
-    "sliding_attention",
-    "full_attention",
-    "sliding_attention",
-    "sliding_attention",
-    "full_attention"
-  ],
-  "local_attention": 128,
-  "max_position_embeddings": 8192,
-  "mlp_bias": false,
-  "mlp_dropout": 0.0,
-  "model_type": "modernbert",
-  "norm_bias": false,
-  "norm_eps": 1e-05,
-  "num_attention_heads": 12,
-  "num_hidden_layers": 22,
-  "pad_token_id": 5,
-  "position_embedding_type": "absolute",
-  "repad_logits_with_grad": false,
-  "rope_parameters": {
-    "full_attention": {
-      "rope_theta": 160000.0,
-      "rope_type": "default"
-    },
-    "sliding_attention": {
-      "rope_theta": 10000.0,
-      "rope_type": "default"
-    }
+    "NONE": 0,
+    "ENTAILMENT": 1
   },
-  "sep_token_id": 50282,
-  "sparse_pred_ignore_index": -100,
-  "sparse_prediction": false,
-  "task": "nli-multitask",
   "tasks": [
     "text-classification",
     "similarity-regression"
   ],
-  "tie_word_embeddings": true,
-  "transformers_version": "5.0.0",
-  "vocab_size": 50368
-}
+  "hidden_size": 768,
+  "vocab_size": 50265,
+  "max_position_embeddings": 512
+}

model.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a131c8f663ba9bca7f73496ba9d80dde5f8b6c3279720a1967509893f9600cac
+size 608422621

optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:026bdc7c6be548208162034d7f13016dda0ccf4aa79222902e3d1fa5f557afff
+size 1215859851

rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2c12e79b9e5dc02c2f4a016b1026f5987487dac8577eddac271f1a988a197d88
+size 14645

scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:defb0eccaec2bd9af1b529c7d9102bc08561dd7771db695d99253463a8577212
+size 1465

trainer_state.json

@@ -0,0 +1,311 @@
+{
+  "best_global_step": 1628,
+  "best_metric": 0.964,
+  "best_model_checkpoint": "model_tasks/checkpoint-1628",
+  "epoch": 4.0,
+  "eval_steps": 500,
+  "global_step": 1628,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.12285012285012285,
+      "grad_norm": 58.800106048583984,
+      "learning_rate": 4.803921568627452e-06,
+      "loss": 13.935902099609375,
+      "step": 50
+    },
+    {
+      "epoch": 0.2457002457002457,
+      "grad_norm": 20.866506576538086,
+      "learning_rate": 9.705882352941177e-06,
+      "loss": 2.0049127197265624,
+      "step": 100
+    },
+    {
+      "epoch": 0.36855036855036855,
+      "grad_norm": 47.32191467285156,
+      "learning_rate": 1.4607843137254903e-05,
+      "loss": 1.2294308471679687,
+      "step": 150
+    },
+    {
+      "epoch": 0.4914004914004914,
+      "grad_norm": 22.775053024291992,
+      "learning_rate": 1.950980392156863e-05,
+      "loss": 0.9614788818359375,
+      "step": 200
+    },
+    {
+      "epoch": 0.6142506142506142,
+      "grad_norm": 20.26030731201172,
+      "learning_rate": 1.9970207816945425e-05,
+      "loss": 0.7626658630371094,
+      "step": 250
+    },
+    {
+      "epoch": 0.7371007371007371,
+      "grad_norm": 15.181203842163086,
+      "learning_rate": 1.9867450374363293e-05,
+      "loss": 0.7171074676513672,
+      "step": 300
+    },
+    {
+      "epoch": 0.85995085995086,
+      "grad_norm": 30.844865798950195,
+      "learning_rate": 1.9692115492385916e-05,
+      "loss": 0.6501062774658203,
+      "step": 350
+    },
+    {
+      "epoch": 0.9828009828009828,
+      "grad_norm": 9.524636268615723,
+      "learning_rate": 1.944549280068335e-05,
+      "loss": 0.5243024063110352,
+      "step": 400
+    },
+    {
+      "epoch": 1.0,
+      "eval_cls_accuracy": 0.902,
+      "eval_cls_f1": 0.9019525450317953,
+      "eval_loss": 0.48792028427124023,
+      "eval_runtime": 2.3223,
+      "eval_samples_per_second": 215.305,
+      "eval_sim_mae": 0.366321861743927,
+      "eval_steps_per_second": 6.89,
+      "step": 407
+    },
+    {
+      "epoch": 1.1056511056511056,
+      "grad_norm": 12.419134140014648,
+      "learning_rate": 1.9129396267685277e-05,
+      "loss": 0.41418304443359377,
+      "step": 450
+    },
+    {
+      "epoch": 1.2285012285012284,
+      "grad_norm": 18.077287673950195,
+      "learning_rate": 1.874615085841268e-05,
+      "loss": 0.3944705581665039,
+      "step": 500
+    },
+    {
+      "epoch": 1.3513513513513513,
+      "grad_norm": 8.001004219055176,
+      "learning_rate": 1.829857543380833e-05,
+      "loss": 0.42740669250488283,
+      "step": 550
+    },
+    {
+      "epoch": 1.4742014742014742,
+      "grad_norm": 20.04601287841797,
+      "learning_rate": 1.7789962017345527e-05,
+      "loss": 0.389661865234375,
+      "step": 600
+    },
+    {
+      "epoch": 1.597051597051597,
+      "grad_norm": 6.977843284606934,
+      "learning_rate": 1.7224051581414144e-05,
+      "loss": 0.38964271545410156,
+      "step": 650
+    },
+    {
+      "epoch": 1.71990171990172,
+      "grad_norm": 7.663310527801514,
+      "learning_rate": 1.6605006531580833e-05,
+      "loss": 0.35614356994628904,
+      "step": 700
+    },
+    {
+      "epoch": 1.8427518427518428,
+      "grad_norm": 12.420454978942871,
+      "learning_rate": 1.593738009110821e-05,
+      "loss": 0.320499153137207,
+      "step": 750
+    },
+    {
+      "epoch": 1.9656019656019657,
+      "grad_norm": 13.657429695129395,
+      "learning_rate": 1.5226082810917096e-05,
+      "loss": 0.3286134719848633,
+      "step": 800
+    },
+    {
+      "epoch": 2.0,
+      "eval_cls_accuracy": 0.946,
+      "eval_cls_f1": 0.9459513562205986,
+      "eval_loss": 0.31798821687698364,
+      "eval_runtime": 2.3034,
+      "eval_samples_per_second": 217.067,
+      "eval_sim_mae": 0.2837124764919281,
+      "eval_steps_per_second": 6.946,
+      "step": 814
+    },
+    {
+      "epoch": 2.0884520884520885,
+      "grad_norm": 3.6505086421966553,
+      "learning_rate": 1.4476346451319078e-05,
+      "loss": 0.2602015495300293,
+      "step": 850
+    },
+    {
+      "epoch": 2.211302211302211,
+      "grad_norm": 7.000202655792236,
+      "learning_rate": 1.3693685501177709e-05,
+      "loss": 0.20756610870361328,
+      "step": 900
+    },
+    {
+      "epoch": 2.3341523341523343,
+      "grad_norm": 8.487826347351074,
+      "learning_rate": 1.2883856617534158e-05,
+      "loss": 0.2469894790649414,
+      "step": 950
+    },
+    {
+      "epoch": 2.457002457002457,
+      "grad_norm": 13.440855979919434,
+      "learning_rate": 1.2052816284028395e-05,
+      "loss": 0.20867227554321288,
+      "step": 1000
+    },
+    {
+      "epoch": 2.57985257985258,
+      "grad_norm": 7.461733341217041,
+      "learning_rate": 1.1206676999548351e-05,
+      "loss": 0.21205909729003905,
+      "step": 1050
+    },
+    {
+      "epoch": 2.7027027027027026,
+      "grad_norm": 15.598981857299805,
+      "learning_rate": 1.0351662319349887e-05,
+      "loss": 0.21189189910888673,
+      "step": 1100
+    },
+    {
+      "epoch": 2.8255528255528253,
+      "grad_norm": 5.610878944396973,
+      "learning_rate": 9.494061079330812e-06,
+      "loss": 0.207319278717041,
+      "step": 1150
+    },
+    {
+      "epoch": 2.9484029484029484,
+      "grad_norm": 11.040838241577148,
+      "learning_rate": 8.640181140150297e-06,
+      "loss": 0.17402687072753906,
+      "step": 1200
+    },
+    {
+      "epoch": 3.0,
+      "eval_cls_accuracy": 0.958,
+      "eval_cls_f1": 0.9579917663862118,
+      "eval_loss": 0.28256744146347046,
+      "eval_runtime": 2.3027,
+      "eval_samples_per_second": 217.134,
+      "eval_sim_mae": 0.25911375880241394,
+      "eval_steps_per_second": 6.948,
+      "step": 1221
+    },
+    {
+      "epoch": 3.0712530712530715,
+      "grad_norm": 6.649824619293213,
+      "learning_rate": 7.79630299141665e-06,
+      "loss": 0.1433491039276123,
+      "step": 1250
+    },
+    {
+      "epoch": 3.194103194103194,
+      "grad_norm": 4.330808639526367,
+      "learning_rate": 6.968633557195666e-06,
+      "loss": 0.11451138496398926,
+      "step": 1300
+    },
+    {
+      "epoch": 3.3169533169533167,
+      "grad_norm": 25.754131317138672,
+      "learning_rate": 6.163260542611005e-06,
+      "loss": 0.12067122459411621,
+      "step": 1350
+    },
+    {
+      "epoch": 3.43980343980344,
+      "grad_norm": 3.0961501598358154,
+      "learning_rate": 5.386107657328183e-06,
+      "loss": 0.09894429206848145,
+      "step": 1400
+    },
+    {
+      "epoch": 3.562653562653563,
+      "grad_norm": 2.771333932876587,
+      "learning_rate": 4.642891045264073e-06,
+      "loss": 0.08938132286071777,
+      "step": 1450
+    },
+    {
+      "epoch": 3.6855036855036856,
+      "grad_norm": 3.6692490577697754,
+      "learning_rate": 3.939077240991749e-06,
+      "loss": 0.10362943649291992,
+      "step": 1500
+    },
+    {
+      "epoch": 3.808353808353808,
+      "grad_norm": 2.1008787155151367,
+      "learning_rate": 3.279842962081324e-06,
+      "loss": 0.09946866989135743,
+      "step": 1550
+    },
+    {
+      "epoch": 3.9312039312039313,
+      "grad_norm": 1.3774856328964233,
+      "learning_rate": 2.670037033113577e-06,
+      "loss": 0.13387378692626953,
+      "step": 1600
+    },
+    {
+      "epoch": 4.0,
+      "eval_cls_accuracy": 0.964,
+      "eval_cls_f1": 0.964,
+      "eval_loss": 0.25317928194999695,
+      "eval_runtime": 2.305,
+      "eval_samples_per_second": 216.923,
+      "eval_sim_mae": 0.2385944277048111,
+      "eval_steps_per_second": 6.942,
+      "step": 1628
+    }
+  ],
+  "logging_steps": 50,
+  "max_steps": 2035,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 5,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "EarlyStoppingCallback": {
+      "args": {
+        "early_stopping_patience": 3,
+        "early_stopping_threshold": 0.0
+      },
+      "attributes": {
+        "early_stopping_patience_counter": 0
+      }
+    },
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 0.0,
+  "train_batch_size": 16,
+  "trial_name": null,
+  "trial_params": null
+}