End of training

README.md

@@ -2,7 +2,7 @@
 tags:
 - generated_from_trainer
 datasets:
-- category_cleaning
+- Displate/category_cleaning
 model-index:
 - name: category_cleaning
   results: []
@@ -13,7 +13,9 @@ should probably proofread and complete it, then remove this comment. -->
 
 # category_cleaning
 
-This model was trained from scratch on the category_cleaning dataset.
+This model was trained from scratch on the Displate/category_cleaning dataset.
+It achieves the following results on the evaluation set:
+- Loss: 4.5536
 
 ## Model description
 

all_results.json

@@ -1,11 +1,11 @@
 {
-    "epoch": 20.0,
-    "eval_loss": 2.79219126701355,
-    "eval_runtime": 0.1465,
-    "eval_samples_per_second": 81.924,
-    "eval_steps_per_second": 6.827,
-    "train_loss": 0.8449213564395904,
-    "train_runtime": 26.3276,
-    "train_samples_per_second": 56.974,
-    "train_steps_per_second": 1.519
+    "epoch": 3.0,
+    "eval_loss": 4.553551197052002,
+    "eval_runtime": 33.5078,
+    "eval_samples_per_second": 124.747,
+    "eval_steps_per_second": 1.97,
+    "train_loss": 2.093686363324016,
+    "train_runtime": 1348.3251,
+    "train_samples_per_second": 55.8,
+    "train_steps_per_second": 0.872
 }

eval_results.json

@@ -1,7 +1,7 @@
 {
-    "epoch": 20.0,
-    "eval_loss": 2.79219126701355,
-    "eval_runtime": 0.1465,
-    "eval_samples_per_second": 81.924,
-    "eval_steps_per_second": 6.827
+    "epoch": 3.0,
+    "eval_loss": 4.553551197052002,
+    "eval_runtime": 33.5078,
+    "eval_samples_per_second": 124.747,
+    "eval_steps_per_second": 1.97
 }

runs/Mar08_14-17-34_9da45c1bcf23/events.out.tfevents.1709908851.9da45c1bcf23.76705.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bd8c02b67c733ee74474f5aa428953d53435e1201306c9c269250bb86c8aff00
+size 359

train_results.json

@@ -1,7 +1,7 @@
 {
-    "epoch": 20.0,
-    "train_loss": 0.8449213564395904,
-    "train_runtime": 26.3276,
-    "train_samples_per_second": 56.974,
-    "train_steps_per_second": 1.519
+    "epoch": 3.0,
+    "train_loss": 2.093686363324016,
+    "train_runtime": 1348.3251,
+    "train_samples_per_second": 55.8,
+    "train_steps_per_second": 0.872
 }

trainer_state.json

@@ -1,85 +1,106 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 20.0,
+  "epoch": 3.0,
   "eval_steps": 500,
-  "global_step": 40,
+  "global_step": 1176,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 2.5,
-      "grad_norm": 37.25093460083008,
-      "learning_rate": 4.375e-05,
-      "loss": 2.2626,
-      "step": 5
+      "epoch": 0.26,
+      "grad_norm": 23.253705978393555,
+      "learning_rate": 4.5748299319727895e-05,
+      "loss": 3.0173,
+      "step": 100
     },
     {
-      "epoch": 5.0,
-      "grad_norm": 49.72227096557617,
-      "learning_rate": 3.7500000000000003e-05,
-      "loss": 0.8291,
-      "step": 10
+      "epoch": 0.51,
+      "grad_norm": 24.05482292175293,
+      "learning_rate": 4.149659863945579e-05,
+      "loss": 2.8146,
+      "step": 200
     },
     {
-      "epoch": 7.5,
-      "grad_norm": 16.81373405456543,
-      "learning_rate": 3.125e-05,
-      "loss": 0.8602,
-      "step": 15
+      "epoch": 0.77,
+      "grad_norm": 19.086837768554688,
+      "learning_rate": 3.724489795918368e-05,
+      "loss": 2.7177,
+      "step": 300
     },
     {
-      "epoch": 10.0,
-      "grad_norm": 7.272302627563477,
-      "learning_rate": 2.5e-05,
-      "loss": 0.5567,
-      "step": 20
+      "epoch": 1.02,
+      "grad_norm": 19.842321395874023,
+      "learning_rate": 3.2993197278911564e-05,
+      "loss": 2.6338,
+      "step": 400
     },
     {
-      "epoch": 12.5,
-      "grad_norm": 9.128313064575195,
-      "learning_rate": 1.8750000000000002e-05,
-      "loss": 0.7576,
-      "step": 25
+      "epoch": 1.28,
+      "grad_norm": 25.054931640625,
+      "learning_rate": 2.8741496598639456e-05,
+      "loss": 2.102,
+      "step": 500
     },
     {
-      "epoch": 15.0,
-      "grad_norm": 14.296534538269043,
-      "learning_rate": 1.25e-05,
-      "loss": 0.4636,
-      "step": 30
+      "epoch": 1.53,
+      "grad_norm": 22.865522384643555,
+      "learning_rate": 2.448979591836735e-05,
+      "loss": 2.1015,
+      "step": 600
     },
     {
-      "epoch": 17.5,
-      "grad_norm": 5.532127380371094,
-      "learning_rate": 6.25e-06,
-      "loss": 0.6616,
-      "step": 35
+      "epoch": 1.79,
+      "grad_norm": 25.445798873901367,
+      "learning_rate": 2.023809523809524e-05,
+      "loss": 2.0691,
+      "step": 700
     },
     {
-      "epoch": 20.0,
-      "grad_norm": 0.8484093546867371,
-      "learning_rate": 0.0,
-      "loss": 0.3679,
-      "step": 40
+      "epoch": 2.04,
+      "grad_norm": 22.133098602294922,
+      "learning_rate": 1.5986394557823133e-05,
+      "loss": 1.9422,
+      "step": 800
     },
     {
-      "epoch": 20.0,
-      "step": 40,
-      "total_flos": 87250496949000.0,
-      "train_loss": 0.8449213564395904,
-      "train_runtime": 26.3276,
-      "train_samples_per_second": 56.974,
-      "train_steps_per_second": 1.519
+      "epoch": 2.3,
+      "grad_norm": 31.207412719726562,
+      "learning_rate": 1.1734693877551021e-05,
+      "loss": 1.4405,
+      "step": 900
+    },
+    {
+      "epoch": 2.55,
+      "grad_norm": 19.457901000976562,
+      "learning_rate": 7.482993197278912e-06,
+      "loss": 1.3896,
+      "step": 1000
+    },
+    {
+      "epoch": 2.81,
+      "grad_norm": 20.708255767822266,
+      "learning_rate": 3.231292517006803e-06,
+      "loss": 1.3735,
+      "step": 1100
+    },
+    {
+      "epoch": 3.0,
+      "step": 1176,
+      "total_flos": 4376310425967942.0,
+      "train_loss": 2.093686363324016,
+      "train_runtime": 1348.3251,
+      "train_samples_per_second": 55.8,
+      "train_steps_per_second": 0.872
     }
   ],
-  "logging_steps": 5,
-  "max_steps": 40,
+  "logging_steps": 100,
+  "max_steps": 1176,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 20,
+  "num_train_epochs": 3,
   "save_steps": 500,
-  "total_flos": 87250496949000.0,
+  "total_flos": 4376310425967942.0,
   "train_batch_size": 64,
   "trial_name": null,
   "trial_params": null