End of training

README.md

@@ -14,9 +14,9 @@ should probably proofread and complete it, then remove this comment. -->
 
 # TAPT_data-V2_Bioformer-16L_LR-0.0005
 
-This model is a fine-tuned version of [bioformers/bioformer-16L](https://huggingface.co/bioformers/bioformer-16L) on an unknown dataset.
+This model is a fine-tuned version of [bioformers/bioformer-16L](https://huggingface.co/bioformers/bioformer-16L) on the Mardiyyah/TAPT_data_V2_split dataset.
 It achieves the following results on the evaluation set:
-- Loss: 3.4749
+- Loss: 3.4623
 
 ## Model description
 

all_results.json

@@ -0,0 +1,15 @@
+{
+    "epoch": 10.0,
+    "eval_loss": 3.4622573852539062,
+    "eval_runtime": 2.2596,
+    "eval_samples": 1945,
+    "eval_samples_per_second": 860.769,
+    "eval_steps_per_second": 53.992,
+    "perplexity": 31.888880815596,
+    "total_flos": 1536221652235776.0,
+    "train_loss": 2.5605053229872228,
+    "train_runtime": 388.3929,
+    "train_samples": 9733,
+    "train_samples_per_second": 250.597,
+    "train_steps_per_second": 15.68
+}

eval_results.json

@@ -0,0 +1,15 @@
+{
+    "epoch": 10.0,
+    "eval_loss": 3.4622573852539062,
+    "eval_runtime": 2.2596,
+    "eval_samples": 1945,
+    "eval_samples_per_second": 860.769,
+    "eval_steps_per_second": 53.992,
+    "perplexity": 31.888880815596,
+    "total_flos": 1536221652235776.0,
+    "train_loss": 2.5605053229872228,
+    "train_runtime": 388.3929,
+    "train_samples": 9733,
+    "train_samples_per_second": 250.597,
+    "train_steps_per_second": 15.68
+}

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 10.0,
+    "total_flos": 1536221652235776.0,
+    "train_loss": 2.5605053229872228,
+    "train_runtime": 388.3929,
+    "train_samples": 9733,
+    "train_samples_per_second": 250.597,
+    "train_steps_per_second": 15.68
+}

trainer_state.json

@@ -0,0 +1,192 @@
+{
+  "best_metric": 3.474938154220581,
+  "best_model_checkpoint": "./TAPT_data-V2_Bioformer-16L_LR-0.0005/checkpoint-6090",
+  "epoch": 10.0,
+  "eval_steps": 500,
+  "global_step": 6090,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "grad_norm": 49.082210540771484,
+      "learning_rate": 0.00047877358490566036,
+      "loss": 2.9555,
+      "step": 609
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 3.884225845336914,
+      "eval_runtime": 2.2953,
+      "eval_samples_per_second": 847.372,
+      "eval_steps_per_second": 53.151,
+      "step": 609
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 29.47579002380371,
+      "learning_rate": 0.0004255765199161425,
+      "loss": 3.4591,
+      "step": 1218
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 4.176673412322998,
+      "eval_runtime": 2.3902,
+      "eval_samples_per_second": 813.746,
+      "eval_steps_per_second": 51.042,
+      "step": 1218
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 20.8883113861084,
+      "learning_rate": 0.00037237945492662474,
+      "loss": 3.3266,
+      "step": 1827
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 4.201623439788818,
+      "eval_runtime": 2.3441,
+      "eval_samples_per_second": 829.735,
+      "eval_steps_per_second": 52.045,
+      "step": 1827
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 20.087749481201172,
+      "learning_rate": 0.0003191823899371069,
+      "loss": 3.0264,
+      "step": 2436
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 4.025837421417236,
+      "eval_runtime": 2.1116,
+      "eval_samples_per_second": 921.11,
+      "eval_steps_per_second": 57.777,
+      "step": 2436
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 24.700023651123047,
+      "learning_rate": 0.00026598532494758907,
+      "loss": 2.7402,
+      "step": 3045
+    },
+    {
+      "epoch": 5.0,
+      "eval_loss": 3.8213584423065186,
+      "eval_runtime": 2.3843,
+      "eval_samples_per_second": 815.767,
+      "eval_steps_per_second": 51.169,
+      "step": 3045
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 21.42559051513672,
+      "learning_rate": 0.00021278825995807126,
+      "loss": 2.4781,
+      "step": 3654
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 3.810668706893921,
+      "eval_runtime": 2.3336,
+      "eval_samples_per_second": 833.486,
+      "eval_steps_per_second": 52.28,
+      "step": 3654
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 20.396045684814453,
+      "learning_rate": 0.00015959119496855345,
+      "loss": 2.2409,
+      "step": 4263
+    },
+    {
+      "epoch": 7.0,
+      "eval_loss": 3.685856819152832,
+      "eval_runtime": 2.1403,
+      "eval_samples_per_second": 908.742,
+      "eval_steps_per_second": 57.001,
+      "step": 4263
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 18.948102951049805,
+      "learning_rate": 0.00010639412997903563,
+      "loss": 1.9816,
+      "step": 4872
+    },
+    {
+      "epoch": 8.0,
+      "eval_loss": 3.6025681495666504,
+      "eval_runtime": 2.1051,
+      "eval_samples_per_second": 923.932,
+      "eval_steps_per_second": 57.954,
+      "step": 4872
+    },
+    {
+      "epoch": 9.0,
+      "grad_norm": 26.760536193847656,
+      "learning_rate": 5.3197064989517815e-05,
+      "loss": 1.7825,
+      "step": 5481
+    },
+    {
+      "epoch": 9.0,
+      "eval_loss": 3.4999165534973145,
+      "eval_runtime": 2.0996,
+      "eval_samples_per_second": 926.384,
+      "eval_steps_per_second": 58.107,
+      "step": 5481
+    },
+    {
+      "epoch": 10.0,
+      "grad_norm": 25.620874404907227,
+      "learning_rate": 0.0,
+      "loss": 1.6141,
+      "step": 6090
+    },
+    {
+      "epoch": 10.0,
+      "eval_loss": 3.474938154220581,
+      "eval_runtime": 2.2881,
+      "eval_samples_per_second": 850.053,
+      "eval_steps_per_second": 53.32,
+      "step": 6090
+    },
+    {
+      "epoch": 10.0,
+      "step": 6090,
+      "total_flos": 1536221652235776.0,
+      "train_loss": 2.5605053229872228,
+      "train_runtime": 388.3929,
+      "train_samples_per_second": 250.597,
+      "train_steps_per_second": 15.68
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 6090,
+  "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": 1536221652235776.0,
+  "train_batch_size": 16,
+  "trial_name": null,
+  "trial_params": null
+}