Upload 159 files

README.md

@@ -17,8 +17,8 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [gpt2](https://huggingface.co/gpt2) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 2.5745
-- Accuracy: 0.4842
+- Loss: 3.0447
+- Accuracy: 0.3878
 
 ## Model description
 

all_results.json

@@ -1,16 +1,16 @@
 {
     "epoch": 20.0,
-    "eval_accuracy": 0.4841968067774519,
-    "eval_loss": 2.5744614601135254,
-    "eval_runtime": 0.6172,
-    "eval_samples": 9,
-    "eval_samples_per_second": 14.581,
-    "eval_steps_per_second": 8.101,
-    "perplexity": 13.124247324302939,
-    "total_flos": 1442332016640000.0,
-    "train_loss": 1.6038585939269134,
-    "train_runtime": 546.7087,
-    "train_samples": 138,
-    "train_samples_per_second": 5.048,
-    "train_steps_per_second": 2.524
+    "eval_accuracy": 0.3878299120234604,
+    "eval_loss": 3.044700860977173,
+    "eval_runtime": 0.5419,
+    "eval_samples": 8,
+    "eval_samples_per_second": 14.762,
+    "eval_steps_per_second": 7.381,
+    "perplexity": 21.00374722261463,
+    "total_flos": 1181039984640000.0,
+    "train_loss": 2.0282494394402755,
+    "train_runtime": 450.0172,
+    "train_samples": 113,
+    "train_samples_per_second": 5.022,
+    "train_steps_per_second": 2.533
 }

checkpoint-1000/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f0a3045c4618f75a91e43e2b4079b4f70ddfea71b549212fe00c8bc207fce25d
+oid sha256:0eafae2b6ff425a68e1dff3b58217192755c5552624bec6e0a5a24c2cfeef0d7
 size 497774208

checkpoint-1000/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a5737a324483690f9aadd907a2e48f602e54024c812e42d2a6c0c28878d67400
+oid sha256:da3865fd609302881be01c29801902af8955f2eaf6a500b6eddbdb1ecbccff3c
 size 995642298

checkpoint-1000/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fad09794a0db580c98cf06d0ebd5b66e92d3f86f4f7bd4a728e24f1a05467f0c
+oid sha256:e15521664bcf9955b1a2b66e810668f825d186fef974c4b18a31c7cd19e0437a
 size 14244

checkpoint-1000/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7e4ac6e2694cbb9ef21cdf8af6f8d76becd952779eb03e00001aef172a063804
+oid sha256:56c9dff28a9d4ea03847c10914973912c6d11b9e39b2afb53832948fd687a713
 size 1064

checkpoint-1000/trainer_state.json

@@ -1,7 +1,7 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 14.492753623188406,
+  "epoch": 17.54385964912281,
   "eval_steps": 500,
   "global_step": 1000,
   "is_hyper_param_search": false,
@@ -9,26 +9,26 @@
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 7.246376811594203,
-      "grad_norm": 4.941216468811035,
-      "learning_rate": 3.188405797101449e-05,
-      "loss": 1.9657,
+      "epoch": 8.771929824561404,
+      "grad_norm": 5.5376715660095215,
+      "learning_rate": 2.8070175438596492e-05,
+      "loss": 2.3748,
       "step": 500
     },
     {
-      "epoch": 14.492753623188406,
-      "grad_norm": 4.604403495788574,
-      "learning_rate": 1.3768115942028985e-05,
-      "loss": 1.4853,
+      "epoch": 17.54385964912281,
+      "grad_norm": 5.5406880378723145,
+      "learning_rate": 6.140350877192982e-06,
+      "loss": 1.7925,
       "step": 1000
     }
   ],
   "logging_steps": 500,
-  "max_steps": 1380,
+  "max_steps": 1140,
   "num_input_tokens_seen": 0,
   "num_train_epochs": 20,
   "save_steps": 500,
-  "total_flos": 1045168128000000.0,
+  "total_flos": 1036284198912000.0,
   "train_batch_size": 2,
   "trial_name": null,
   "trial_params": null

checkpoint-1000/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d74f83bec7ea349d722f90da656054543e87196b485a7f973f3b2bf816599956
+oid sha256:cb06b4254316e1ab5e3df832bc95c5eed217d0c6e05ef2ece9717f8db05462b9
 size 4984

checkpoint-500/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:675128bde697f16287cd94a082b72bc459c336abc30693b18347577766156deb
+oid sha256:d6f130cbd8a4515242330284192392edc8462adce4467e1764fab022c447c13a
 size 497774208

checkpoint-500/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6e92d1b34033798c941cd79408ba58778a2e495f100658eb08b65531a075ad4f
+oid sha256:291721543b1bea75dc67b7f4214f7e495d5792dc17f7bbe192b2c5c500d1ce54
 size 995642298

checkpoint-500/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2468835bf602a0fb3a2f44ad85744a106df5db55b9864b0626319bec2d54fb1d
+oid sha256:777137f7b6438e37de19223f30fb87cd1210813798ec473dff745146fb62b610
 size 14244

checkpoint-500/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cf5b518ffefd08bf15650baa38cbbf384a2122ba91aea9d95cd9aa0329fa2678
+oid sha256:d9df7e73f7009fee1832457c31aeab89c9411a8aad568d49bf516ca4a4b45d88
 size 1064

checkpoint-500/trainer_state.json

@@ -1,7 +1,7 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 7.246376811594203,
+  "epoch": 8.771929824561404,
   "eval_steps": 500,
   "global_step": 500,
   "is_hyper_param_search": false,
@@ -9,19 +9,19 @@
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 7.246376811594203,
-      "grad_norm": 4.941216468811035,
-      "learning_rate": 3.188405797101449e-05,
-      "loss": 1.9657,
+      "epoch": 8.771929824561404,
+      "grad_norm": 5.5376715660095215,
+      "learning_rate": 2.8070175438596492e-05,
+      "loss": 2.3748,
       "step": 500
     }
   ],
   "logging_steps": 500,
-  "max_steps": 1380,
+  "max_steps": 1140,
   "num_input_tokens_seen": 0,
   "num_train_epochs": 20,
   "save_steps": 500,
-  "total_flos": 522584064000000.0,
+  "total_flos": 518403391488000.0,
   "train_batch_size": 2,
   "trial_name": null,
   "trial_params": null

checkpoint-500/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d74f83bec7ea349d722f90da656054543e87196b485a7f973f3b2bf816599956
+oid sha256:cb06b4254316e1ab5e3df832bc95c5eed217d0c6e05ef2ece9717f8db05462b9
 size 4984

eval_results.json

@@ -1,10 +1,10 @@
 {
     "epoch": 20.0,
-    "eval_accuracy": 0.4841968067774519,
-    "eval_loss": 2.5744614601135254,
-    "eval_runtime": 0.6172,
-    "eval_samples": 9,
-    "eval_samples_per_second": 14.581,
-    "eval_steps_per_second": 8.101,
-    "perplexity": 13.124247324302939
+    "eval_accuracy": 0.3878299120234604,
+    "eval_loss": 3.044700860977173,
+    "eval_runtime": 0.5419,
+    "eval_samples": 8,
+    "eval_samples_per_second": 14.762,
+    "eval_steps_per_second": 7.381,
+    "perplexity": 21.00374722261463
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:80b827b98d26451ec1300044ca990f9d2c0a2e66d6928b453bb6246c6e355525
+oid sha256:ed42548dc68d3a36c9021a2620c13fbad3926a37f8f73cc0176886c465edf1ad
 size 497774208

train_results.json

@@ -1,9 +1,9 @@
 {
     "epoch": 20.0,
-    "total_flos": 1442332016640000.0,
-    "train_loss": 1.6038585939269134,
-    "train_runtime": 546.7087,
-    "train_samples": 138,
-    "train_samples_per_second": 5.048,
-    "train_steps_per_second": 2.524
+    "total_flos": 1181039984640000.0,
+    "train_loss": 2.0282494394402755,
+    "train_runtime": 450.0172,
+    "train_samples": 113,
+    "train_samples_per_second": 5.022,
+    "train_steps_per_second": 2.533
 }

trainer_state.json

@@ -3,41 +3,41 @@
   "best_model_checkpoint": null,
   "epoch": 20.0,
   "eval_steps": 500,
-  "global_step": 1380,
+  "global_step": 1140,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 7.246376811594203,
-      "grad_norm": 4.941216468811035,
-      "learning_rate": 3.188405797101449e-05,
-      "loss": 1.9657,
+      "epoch": 8.771929824561404,
+      "grad_norm": 5.5376715660095215,
+      "learning_rate": 2.8070175438596492e-05,
+      "loss": 2.3748,
       "step": 500
     },
     {
-      "epoch": 14.492753623188406,
-      "grad_norm": 4.604403495788574,
-      "learning_rate": 1.3768115942028985e-05,
-      "loss": 1.4853,
+      "epoch": 17.54385964912281,
+      "grad_norm": 5.5406880378723145,
+      "learning_rate": 6.140350877192982e-06,
+      "loss": 1.7925,
       "step": 1000
     },
     {
       "epoch": 20.0,
-      "step": 1380,
-      "total_flos": 1442332016640000.0,
-      "train_loss": 1.6038585939269134,
-      "train_runtime": 546.7087,
-      "train_samples_per_second": 5.048,
-      "train_steps_per_second": 2.524
+      "step": 1140,
+      "total_flos": 1181039984640000.0,
+      "train_loss": 2.0282494394402755,
+      "train_runtime": 450.0172,
+      "train_samples_per_second": 5.022,
+      "train_steps_per_second": 2.533
     }
   ],
   "logging_steps": 500,
-  "max_steps": 1380,
+  "max_steps": 1140,
   "num_input_tokens_seen": 0,
   "num_train_epochs": 20,
   "save_steps": 500,
-  "total_flos": 1442332016640000.0,
+  "total_flos": 1181039984640000.0,
   "train_batch_size": 2,
   "trial_name": null,
   "trial_params": null

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d74f83bec7ea349d722f90da656054543e87196b485a7f973f3b2bf816599956
+oid sha256:cb06b4254316e1ab5e3df832bc95c5eed217d0c6e05ef2ece9717f8db05462b9
 size 4984