correct weights

README.md

@@ -5,14 +5,14 @@ metrics:
 - f1
 - accuracy
 model-index:
-- name: final-lr2e-5-bs16
+- name: final-lr2e-5-bs16-fp16-2
   results: []
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
 should probably proofread and complete it, then remove this comment. -->
 
-# final-lr2e-5-bs16
+# final-lr2e-5-bs16-fp16-2
 
 This model is a fine-tuned version of [clincolnoz/LessSexistBERT](https://huggingface.co/clincolnoz/LessSexistBERT) on an unknown dataset.
 It achieves the following results on the evaluation set:

all_results.json

@@ -8,13 +8,13 @@
     "eval_f1_macro": 0.8374201959393394,
     "eval_f1_weighted": 0.8806450852201921,
     "eval_loss": 0.3457934856414795,
-    "eval_runtime": 9.6259,
+    "eval_runtime": 9.6249,
     "eval_samples": 4000,
-    "eval_samples_per_second": 415.545,
-    "eval_steps_per_second": 25.972,
+    "eval_samples_per_second": 415.588,
+    "eval_steps_per_second": 25.974,
     "train_loss": 0.27966182454427085,
-    "train_runtime": 436.5097,
+    "train_runtime": 430.7509,
     "train_samples": 16000,
-    "train_samples_per_second": 109.963,
-    "train_steps_per_second": 6.873
+    "train_samples_per_second": 111.433,
+    "train_steps_per_second": 6.965
 }

eval_results.json

@@ -8,8 +8,8 @@
     "eval_f1_macro": 0.8374201959393394,
     "eval_f1_weighted": 0.8806450852201921,
     "eval_loss": 0.3457934856414795,
-    "eval_runtime": 9.6259,
+    "eval_runtime": 9.6249,
     "eval_samples": 4000,
-    "eval_samples_per_second": 415.545,
-    "eval_steps_per_second": 25.972
+    "eval_samples_per_second": 415.588,
+    "eval_steps_per_second": 25.974
 }

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 3.0,
     "train_loss": 0.27966182454427085,
-    "train_runtime": 436.5097,
+    "train_runtime": 430.7509,
     "train_samples": 16000,
-    "train_samples_per_second": 109.963,
-    "train_steps_per_second": 6.873
+    "train_samples_per_second": 111.433,
+    "train_steps_per_second": 6.965
 }

trainer_state.json

@@ -29,9 +29,9 @@
       "eval_f1_macro": 0.8256385630330887,
       "eval_f1_weighted": 0.8748468477738904,
       "eval_loss": 0.3010723888874054,
-      "eval_runtime": 9.6387,
-      "eval_samples_per_second": 414.993,
-      "eval_steps_per_second": 25.937,
+      "eval_runtime": 9.6596,
+      "eval_samples_per_second": 414.097,
+      "eval_steps_per_second": 25.881,
       "step": 1000
     },
     {
@@ -56,9 +56,9 @@
       "eval_f1_macro": 0.84110534680508,
       "eval_f1_weighted": 0.8848438807534127,
       "eval_loss": 0.3121766746044159,
-      "eval_runtime": 9.6263,
-      "eval_samples_per_second": 415.529,
-      "eval_steps_per_second": 25.971,
+      "eval_runtime": 9.6642,
+      "eval_samples_per_second": 413.897,
+      "eval_steps_per_second": 25.869,
       "step": 2000
     },
     {
@@ -83,9 +83,9 @@
       "eval_f1_macro": 0.8374201959393394,
       "eval_f1_weighted": 0.8806450852201921,
       "eval_loss": 0.3457934856414795,
-      "eval_runtime": 9.6302,
-      "eval_samples_per_second": 415.361,
-      "eval_steps_per_second": 25.96,
+      "eval_runtime": 9.6651,
+      "eval_samples_per_second": 413.86,
+      "eval_steps_per_second": 25.866,
       "step": 3000
     },
     {
@@ -93,9 +93,9 @@
       "step": 3000,
       "total_flos": 1.262933065728e+16,
       "train_loss": 0.27966182454427085,
-      "train_runtime": 436.5097,
-      "train_samples_per_second": 109.963,
-      "train_steps_per_second": 6.873
+      "train_runtime": 430.7509,
+      "train_samples_per_second": 111.433,
+      "train_steps_per_second": 6.965
     }
   ],
   "max_steps": 3000,

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cf156ae74a37a378f8916ca566a842b16575ec3ad5384916e9c69010b34a1ae0
+oid sha256:19635231ad63260728a01f0ae1dced21f343315a5fc1070be2fd708a403c7343
 size 3643