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 [GroNLP/hateBERT](https://huggingface.co/GroNLP/hateBERT) on an unknown dataset.
 It achieves the following results on the evaluation set:

all_results.json

@@ -8,13 +8,13 @@
     "eval_f1_macro": 0.8456692644535684,
     "eval_f1_weighted": 0.8868003346108867,
     "eval_loss": 0.42191165685653687,
-    "eval_runtime": 9.6157,
+    "eval_runtime": 9.5988,
     "eval_samples": 4000,
-    "eval_samples_per_second": 415.988,
-    "eval_steps_per_second": 25.999,
+    "eval_samples_per_second": 416.717,
+    "eval_steps_per_second": 26.045,
     "train_loss": 0.25008199564615885,
-    "train_runtime": 436.7193,
+    "train_runtime": 430.1638,
     "train_samples": 16000,
-    "train_samples_per_second": 109.91,
-    "train_steps_per_second": 6.869
+    "train_samples_per_second": 111.585,
+    "train_steps_per_second": 6.974
 }

eval_results.json

@@ -8,8 +8,8 @@
     "eval_f1_macro": 0.8456692644535684,
     "eval_f1_weighted": 0.8868003346108867,
     "eval_loss": 0.42191165685653687,
-    "eval_runtime": 9.6157,
+    "eval_runtime": 9.5988,
     "eval_samples": 4000,
-    "eval_samples_per_second": 415.988,
-    "eval_steps_per_second": 25.999
+    "eval_samples_per_second": 416.717,
+    "eval_steps_per_second": 26.045
 }

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 3.0,
     "train_loss": 0.25008199564615885,
-    "train_runtime": 436.7193,
+    "train_runtime": 430.1638,
     "train_samples": 16000,
-    "train_samples_per_second": 109.91,
-    "train_steps_per_second": 6.869
+    "train_samples_per_second": 111.585,
+    "train_steps_per_second": 6.974
 }

trainer_state.json

@@ -29,9 +29,9 @@
       "eval_f1_macro": 0.8322651270370958,
       "eval_f1_weighted": 0.8811581651570527,
       "eval_loss": 0.28936144709587097,
-      "eval_runtime": 9.6533,
-      "eval_samples_per_second": 414.366,
-      "eval_steps_per_second": 25.898,
+      "eval_runtime": 9.6197,
+      "eval_samples_per_second": 415.814,
+      "eval_steps_per_second": 25.988,
       "step": 1000
     },
     {
@@ -56,9 +56,9 @@
       "eval_f1_macro": 0.8404912868365435,
       "eval_f1_weighted": 0.883022889500339,
       "eval_loss": 0.3370012938976288,
-      "eval_runtime": 9.6971,
-      "eval_samples_per_second": 412.493,
-      "eval_steps_per_second": 25.781,
+      "eval_runtime": 9.6013,
+      "eval_samples_per_second": 416.608,
+      "eval_steps_per_second": 26.038,
       "step": 2000
     },
     {
@@ -83,9 +83,9 @@
       "eval_f1_macro": 0.8456692644535684,
       "eval_f1_weighted": 0.8868003346108867,
       "eval_loss": 0.42191165685653687,
-      "eval_runtime": 9.677,
-      "eval_samples_per_second": 413.351,
-      "eval_steps_per_second": 25.834,
+      "eval_runtime": 9.6938,
+      "eval_samples_per_second": 412.636,
+      "eval_steps_per_second": 25.79,
       "step": 3000
     },
     {
@@ -93,9 +93,9 @@
       "step": 3000,
       "total_flos": 1.262933065728e+16,
       "train_loss": 0.25008199564615885,
-      "train_runtime": 436.7193,
-      "train_samples_per_second": 109.91,
-      "train_steps_per_second": 6.869
+      "train_runtime": 430.1638,
+      "train_samples_per_second": 111.585,
+      "train_steps_per_second": 6.974
     }
   ],
   "max_steps": 3000,

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2abc00e67321e82c100e98b2c67ea39b1d04e317f8ffe4e881a7a85819f19003
+oid sha256:7e9edea7fbc0c6473a0822fe1b148f4242cc30096aab146072b47a4fdb2ebdca
 size 3579