Training in progress, step 500

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ecd7bb18dc112a2febe521e6b29f5a5b9eec145154d2663152f38803378acdc7
+oid sha256:c523e1752fda46381be0344a31afc01362450b48bbf8705a2c2bdc36cbf5d0d2
 size 268290900

run-14/checkpoint-1000/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2223e80d5ea251e3b1623748a4a86ab2d2981dcdddbeb483ed74ad70ab89e4b1
+oid sha256:7505fa110151143b55727f37c65dd0726b07a6d91b9f547a3cd9a89b9a4bd139
 size 268290900

run-14/checkpoint-1000/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c67385522485781d97f8085ba19a13c52ffe06af5731820d5a35df270a890307
+oid sha256:0b460d9b2ed09db2956bcfca01e1373a79e023c49a63d6704f041504b5cb8207
 size 536643898

run-14/checkpoint-1000/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2ae8803a4cfd93a17b836d411a0cdc283947bea33521d8af530f6f31af83987d
+oid sha256:c35ee7e612988cfbcef57649260f5e638d50749c336bb364c0aaa11036e867c9
 size 14244

run-14/checkpoint-1000/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:045cf4abe065257226fe11ac1a91dfd4c5b953cb2366365ef8bb18b4cf33e734
+oid sha256:72a4307cca951ef6d42c7f184774e8b1259c78b2bf130b8443ec0b62f6dbf689
 size 1064

run-14/checkpoint-1000/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:82074645e30ff5a42a72a9e1d44203575f3a0877ba6b6be0e8a33d15508ff41a
-size 4728
+oid sha256:f48699de53d88da8e9ede6886c1b6802a2f53470436e3cd9627d81b82b771d90
+size 5176

run-14/checkpoint-1500/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:106d03fb619431388a2c6c0fb5cde37e682e671427afbe1afcbf94001d8b7d97
+oid sha256:f5ae625825faead8ed9468c0a643a575c2fee37edaae3c6d29eb222884fec995
 size 268290900

run-14/checkpoint-1500/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c49b8976e5565b04a3c3eeaf8b10a084d4b630d25d75444f5de103bbc5331fd1
+oid sha256:14dddd0e4396f6fd373103aadaccfa6644c9ee4a48ad7bbfb0a8b8bb6cc23624
 size 536643898

run-14/checkpoint-1500/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3e4150b40cdd095f99e6d506db714a6c0a9a316d6b4d021acbfec8d198c96919
+oid sha256:1c10319269a5ccb2640eed78049edda0fca1ea40d757f64b6ab5474fe9e926f9
 size 14244

run-14/checkpoint-1500/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4ecffc9352fe3049303c30dad0aa46b5b31065b1c973c1f20e15bae4914342d6
-size 4728
+oid sha256:f48699de53d88da8e9ede6886c1b6802a2f53470436e3cd9627d81b82b771d90
+size 5176

run-14/checkpoint-2000/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:71b04ce6835dfbae1a886a5e8f7d6a9079225e5683e06b62e53cae46bd989874
+oid sha256:6dfe6c2a3ea2ffff417869a8ba2e4be6d3a1c92e548ab2ce525904b64369c3d8
 size 268290900

run-14/checkpoint-2000/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1fcf85e9c24f8950ea868bdf4b135570d2fe137de9a671514b3d36ac10fa76bb
+oid sha256:202f5c67c4021e77b6164ac1aa4a75c46646815c92e2ee6d6677bc95a93c73df
 size 536643898

run-14/checkpoint-2000/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5613af56824b89d2251abc079d69f51d9497a128212db6eb459f4c957e78a29b
+oid sha256:37ccc14cc902672c3d8e770c2a2ce3a532b454981f3118cb18a09274102a99fc
 size 14244

run-14/checkpoint-2000/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4ecffc9352fe3049303c30dad0aa46b5b31065b1c973c1f20e15bae4914342d6
-size 4728
+oid sha256:f48699de53d88da8e9ede6886c1b6802a2f53470436e3cd9627d81b82b771d90
+size 5176

run-14/checkpoint-500/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:85ea48c883bea8fecb5235623d8511ac7cee329d148b8788b1a6dd1fddd8256b
+oid sha256:a49665e86cf14fd4d2f3fa4f9df3a01cfe101697fe62edafb2da592098ecdf79
 size 268290900

run-14/checkpoint-500/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8a914e2ea3ef79ac435ce72f2da9edb91f9344490fc3ee879a135fcdd3548ab6
+oid sha256:c5b610b8ae9d31d4e7275bde792df5fb6cfb5bd8f6078f83ac6fbc41ae79253a
 size 536643898

run-14/checkpoint-500/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f8174c1d6ac2627de77161655e4e20ef37d9f2235e2e7c4adc0da0d4e0e14b6d
+oid sha256:b9fb2a0dfb1b0ccef3590fe01e9bd16b6db86f247cc9c7e77290c217a53bac20
 size 14244

run-14/checkpoint-500/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:90dc4637e972cc69c745eebddd8a7560dca27d2318df3e23f8e145abbf236536
+oid sha256:98041bd7cae455426e290a1a0ee683bd5dd30893f7451fec3a464ae8995b17e4
 size 1064

run-14/checkpoint-500/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:82074645e30ff5a42a72a9e1d44203575f3a0877ba6b6be0e8a33d15508ff41a
-size 4728
+oid sha256:f48699de53d88da8e9ede6886c1b6802a2f53470436e3cd9627d81b82b771d90
+size 5176

run-15/checkpoint-1000/config.json

@@ -326,6 +326,6 @@
   "sinusoidal_pos_embds": false,
   "tie_weights_": true,
   "torch_dtype": "float32",
-  "transformers_version": "4.37.2",
+  "transformers_version": "4.41.1",
   "vocab_size": 30522
 }

run-15/checkpoint-1000/tokenizer.json

@@ -1,11 +1,6 @@
 {
   "version": "1.0",
-  "truncation": {
-    "direction": "Right",
-    "max_length": 512,
-    "strategy": "LongestFirst",
-    "stride": 0
-  },
+  "truncation": null,
   "padding": null,
   "added_tokens": [
     {

run-15/checkpoint-1000/trainer_state.json

@@ -10,55 +10,69 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.67,
-      "eval_loss": 0.414095014333725,
-      "eval_runtime": 1.3801,
-      "eval_samples_per_second": 2246.155,
-      "eval_steps_per_second": 47.097,
+      "eval_accuracy": 0.62,
+      "eval_loss": 0.249518021941185,
+      "eval_runtime": 1.3697,
+      "eval_samples_per_second": 2263.294,
+      "eval_steps_per_second": 47.456,
       "step": 318
     },
     {
-      "epoch": 1.57,
-      "learning_rate": 1.606918238993711e-05,
-      "loss": 0.6462,
+      "epoch": 1.5723270440251573,
+      "grad_norm": 0.6149903535842896,
+      "learning_rate": 1.650593990216632e-05,
+      "loss": 0.3991,
       "step": 500
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.8393548387096774,
-      "eval_loss": 0.1437353640794754,
-      "eval_runtime": 1.3827,
-      "eval_samples_per_second": 2242.01,
-      "eval_steps_per_second": 47.01,
+      "eval_accuracy": 0.8422580645161291,
+      "eval_loss": 0.11037396639585495,
+      "eval_runtime": 1.3637,
+      "eval_samples_per_second": 2273.273,
+      "eval_steps_per_second": 47.665,
       "step": 636
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.8964516129032258,
-      "eval_loss": 0.07364548742771149,
-      "eval_runtime": 1.3816,
-      "eval_samples_per_second": 2243.785,
-      "eval_steps_per_second": 47.047,
+      "eval_accuracy": 0.8883870967741936,
+      "eval_loss": 0.0677555724978447,
+      "eval_runtime": 1.3669,
+      "eval_samples_per_second": 2267.902,
+      "eval_steps_per_second": 47.553,
       "step": 954
     },
     {
-      "epoch": 3.14,
-      "learning_rate": 1.2138364779874214e-05,
-      "loss": 0.1675,
+      "epoch": 3.1446540880503147,
+      "grad_norm": 0.5547620058059692,
+      "learning_rate": 1.3011879804332637e-05,
+      "loss": 0.1288,
       "step": 1000
     }
   ],
   "logging_steps": 500,
-  "max_steps": 2544,
+  "max_steps": 2862,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 8,
+  "num_train_epochs": 9,
   "save_steps": 500,
-  "total_flos": 330815748742752.0,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 319943755193412.0,
   "train_batch_size": 48,
   "trial_name": null,
   "trial_params": {
-    "alpha": 0.3292104451728688,
-    "num_train_epochs": 8,
-    "temperature": 2
+    "alpha": 0.9970515290608379,
+    "num_train_epochs": 9,
+    "temperature": 4
   }
 }

run-15/checkpoint-1500/config.json

@@ -326,6 +326,6 @@
   "sinusoidal_pos_embds": false,
   "tie_weights_": true,
   "torch_dtype": "float32",
-  "transformers_version": "4.37.2",
+  "transformers_version": "4.41.1",
   "vocab_size": 30522
 }

run-15/checkpoint-1500/tokenizer.json

@@ -1,11 +1,6 @@
 {
   "version": "1.0",
-  "truncation": {
-    "direction": "Right",
-    "max_length": 512,
-    "strategy": "LongestFirst",
-    "stride": 0
-  },
+  "truncation": null,
   "padding": null,
   "added_tokens": [
     {

run-15/checkpoint-1500/trainer_state.json

@@ -10,70 +10,85 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.67,
-      "eval_loss": 0.414095014333725,
-      "eval_runtime": 1.3801,
-      "eval_samples_per_second": 2246.155,
-      "eval_steps_per_second": 47.097,
+      "eval_accuracy": 0.62,
+      "eval_loss": 0.249518021941185,
+      "eval_runtime": 1.3697,
+      "eval_samples_per_second": 2263.294,
+      "eval_steps_per_second": 47.456,
       "step": 318
     },
     {
-      "epoch": 1.57,
-      "learning_rate": 1.606918238993711e-05,
-      "loss": 0.6462,
+      "epoch": 1.5723270440251573,
+      "grad_norm": 0.6149903535842896,
+      "learning_rate": 1.650593990216632e-05,
+      "loss": 0.3991,
       "step": 500
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.8393548387096774,
-      "eval_loss": 0.1437353640794754,
-      "eval_runtime": 1.3827,
-      "eval_samples_per_second": 2242.01,
-      "eval_steps_per_second": 47.01,
+      "eval_accuracy": 0.8422580645161291,
+      "eval_loss": 0.11037396639585495,
+      "eval_runtime": 1.3637,
+      "eval_samples_per_second": 2273.273,
+      "eval_steps_per_second": 47.665,
       "step": 636
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.8964516129032258,
-      "eval_loss": 0.07364548742771149,
-      "eval_runtime": 1.3816,
-      "eval_samples_per_second": 2243.785,
-      "eval_steps_per_second": 47.047,
+      "eval_accuracy": 0.8883870967741936,
+      "eval_loss": 0.0677555724978447,
+      "eval_runtime": 1.3669,
+      "eval_samples_per_second": 2267.902,
+      "eval_steps_per_second": 47.553,
       "step": 954
     },
     {
-      "epoch": 3.14,
-      "learning_rate": 1.2138364779874214e-05,
-      "loss": 0.1675,
+      "epoch": 3.1446540880503147,
+      "grad_norm": 0.5547620058059692,
+      "learning_rate": 1.3011879804332637e-05,
+      "loss": 0.1288,
       "step": 1000
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.917741935483871,
-      "eval_loss": 0.05248909071087837,
-      "eval_runtime": 1.3876,
-      "eval_samples_per_second": 2234.053,
-      "eval_steps_per_second": 46.843,
+      "eval_accuracy": 0.902258064516129,
+      "eval_loss": 0.04977014288306236,
+      "eval_runtime": 1.3624,
+      "eval_samples_per_second": 2275.356,
+      "eval_steps_per_second": 47.709,
       "step": 1272
     },
     {
-      "epoch": 4.72,
-      "learning_rate": 8.207547169811321e-06,
-      "loss": 0.0853,
+      "epoch": 4.716981132075472,
+      "grad_norm": 0.3585783839225769,
+      "learning_rate": 9.517819706498952e-06,
+      "loss": 0.076,
       "step": 1500
     }
   ],
   "logging_steps": 500,
-  "max_steps": 2544,
+  "max_steps": 2862,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 8,
+  "num_train_epochs": 9,
   "save_steps": 500,
-  "total_flos": 461134918951140.0,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 450371359983132.0,
   "train_batch_size": 48,
   "trial_name": null,
   "trial_params": {
-    "alpha": 0.3292104451728688,
-    "num_train_epochs": 8,
-    "temperature": 2
+    "alpha": 0.9970515290608379,
+    "num_train_epochs": 9,
+    "temperature": 4
   }
 }

run-15/checkpoint-2000/config.json

@@ -326,6 +326,6 @@
   "sinusoidal_pos_embds": false,
   "tie_weights_": true,
   "torch_dtype": "float32",
-  "transformers_version": "4.37.2",
+  "transformers_version": "4.41.1",
   "vocab_size": 30522
 }

run-15/checkpoint-2000/tokenizer.json

@@ -1,11 +1,6 @@
 {
   "version": "1.0",
-  "truncation": {
-    "direction": "Right",
-    "max_length": 512,
-    "strategy": "LongestFirst",
-    "stride": 0
-  },
+  "truncation": null,
   "padding": null,
   "added_tokens": [
     {

run-15/checkpoint-2000/trainer_state.json

@@ -10,94 +10,110 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.67,
-      "eval_loss": 0.414095014333725,
-      "eval_runtime": 1.3801,
-      "eval_samples_per_second": 2246.155,
-      "eval_steps_per_second": 47.097,
+      "eval_accuracy": 0.62,
+      "eval_loss": 0.249518021941185,
+      "eval_runtime": 1.3697,
+      "eval_samples_per_second": 2263.294,
+      "eval_steps_per_second": 47.456,
       "step": 318
     },
     {
-      "epoch": 1.57,
-      "learning_rate": 1.606918238993711e-05,
-      "loss": 0.6462,
+      "epoch": 1.5723270440251573,
+      "grad_norm": 0.6149903535842896,
+      "learning_rate": 1.650593990216632e-05,
+      "loss": 0.3991,
       "step": 500
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.8393548387096774,
-      "eval_loss": 0.1437353640794754,
-      "eval_runtime": 1.3827,
-      "eval_samples_per_second": 2242.01,
-      "eval_steps_per_second": 47.01,
+      "eval_accuracy": 0.8422580645161291,
+      "eval_loss": 0.11037396639585495,
+      "eval_runtime": 1.3637,
+      "eval_samples_per_second": 2273.273,
+      "eval_steps_per_second": 47.665,
       "step": 636
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.8964516129032258,
-      "eval_loss": 0.07364548742771149,
-      "eval_runtime": 1.3816,
-      "eval_samples_per_second": 2243.785,
-      "eval_steps_per_second": 47.047,
+      "eval_accuracy": 0.8883870967741936,
+      "eval_loss": 0.0677555724978447,
+      "eval_runtime": 1.3669,
+      "eval_samples_per_second": 2267.902,
+      "eval_steps_per_second": 47.553,
       "step": 954
     },
     {
-      "epoch": 3.14,
-      "learning_rate": 1.2138364779874214e-05,
-      "loss": 0.1675,
+      "epoch": 3.1446540880503147,
+      "grad_norm": 0.5547620058059692,
+      "learning_rate": 1.3011879804332637e-05,
+      "loss": 0.1288,
       "step": 1000
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.917741935483871,
-      "eval_loss": 0.05248909071087837,
-      "eval_runtime": 1.3876,
-      "eval_samples_per_second": 2234.053,
-      "eval_steps_per_second": 46.843,
+      "eval_accuracy": 0.902258064516129,
+      "eval_loss": 0.04977014288306236,
+      "eval_runtime": 1.3624,
+      "eval_samples_per_second": 2275.356,
+      "eval_steps_per_second": 47.709,
       "step": 1272
     },
     {
-      "epoch": 4.72,
-      "learning_rate": 8.207547169811321e-06,
-      "loss": 0.0853,
+      "epoch": 4.716981132075472,
+      "grad_norm": 0.3585783839225769,
+      "learning_rate": 9.517819706498952e-06,
+      "loss": 0.076,
       "step": 1500
     },
     {
       "epoch": 5.0,
-      "eval_accuracy": 0.9274193548387096,
-      "eval_loss": 0.044280897825956345,
-      "eval_runtime": 1.3907,
-      "eval_samples_per_second": 2229.067,
-      "eval_steps_per_second": 46.739,
+      "eval_accuracy": 0.9167741935483871,
+      "eval_loss": 0.039991483092308044,
+      "eval_runtime": 1.3719,
+      "eval_samples_per_second": 2259.57,
+      "eval_steps_per_second": 47.378,
       "step": 1590
     },
     {
       "epoch": 6.0,
-      "eval_accuracy": 0.9293548387096774,
-      "eval_loss": 0.03997669741511345,
-      "eval_runtime": 1.3903,
-      "eval_samples_per_second": 2229.747,
-      "eval_steps_per_second": 46.753,
+      "eval_accuracy": 0.9232258064516129,
+      "eval_loss": 0.034951966255903244,
+      "eval_runtime": 1.3675,
+      "eval_samples_per_second": 2266.904,
+      "eval_steps_per_second": 47.532,
       "step": 1908
     },
     {
-      "epoch": 6.29,
-      "learning_rate": 4.276729559748428e-06,
-      "loss": 0.0651,
+      "epoch": 6.289308176100629,
+      "grad_norm": 0.27236634492874146,
+      "learning_rate": 6.02375960866527e-06,
+      "loss": 0.0585,
       "step": 2000
     }
   ],
   "logging_steps": 500,
-  "max_steps": 2544,
+  "max_steps": 2862,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 8,
+  "num_train_epochs": 9,
   "save_steps": 500,
-  "total_flos": 590095803351264.0,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 579993747211956.0,
   "train_batch_size": 48,
   "trial_name": null,
   "trial_params": {
-    "alpha": 0.3292104451728688,
-    "num_train_epochs": 8,
-    "temperature": 2
+    "alpha": 0.9970515290608379,
+    "num_train_epochs": 9,
+    "temperature": 4
   }
 }

run-15/checkpoint-2500/config.json

@@ -326,6 +326,6 @@
   "sinusoidal_pos_embds": false,
   "tie_weights_": true,
   "torch_dtype": "float32",
-  "transformers_version": "4.37.2",
+  "transformers_version": "4.41.1",
   "vocab_size": 30522
 }

run-15/checkpoint-2500/tokenizer.json

@@ -1,11 +1,6 @@
 {
   "version": "1.0",
-  "truncation": {
-    "direction": "Right",
-    "max_length": 512,
-    "strategy": "LongestFirst",
-    "stride": 0
-  },
+  "truncation": null,
   "padding": null,
   "added_tokens": [
     {

run-15/checkpoint-2500/trainer_state.json

@@ -10,109 +10,126 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.67,
-      "eval_loss": 0.414095014333725,
-      "eval_runtime": 1.3801,
-      "eval_samples_per_second": 2246.155,
-      "eval_steps_per_second": 47.097,
+      "eval_accuracy": 0.62,
+      "eval_loss": 0.249518021941185,
+      "eval_runtime": 1.3697,
+      "eval_samples_per_second": 2263.294,
+      "eval_steps_per_second": 47.456,
       "step": 318
     },
     {
-      "epoch": 1.57,
-      "learning_rate": 1.606918238993711e-05,
-      "loss": 0.6462,
+      "epoch": 1.5723270440251573,
+      "grad_norm": 0.6149903535842896,
+      "learning_rate": 1.650593990216632e-05,
+      "loss": 0.3991,
       "step": 500
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.8393548387096774,
-      "eval_loss": 0.1437353640794754,
-      "eval_runtime": 1.3827,
-      "eval_samples_per_second": 2242.01,
-      "eval_steps_per_second": 47.01,
+      "eval_accuracy": 0.8422580645161291,
+      "eval_loss": 0.11037396639585495,
+      "eval_runtime": 1.3637,
+      "eval_samples_per_second": 2273.273,
+      "eval_steps_per_second": 47.665,
       "step": 636
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.8964516129032258,
-      "eval_loss": 0.07364548742771149,
-      "eval_runtime": 1.3816,
-      "eval_samples_per_second": 2243.785,
-      "eval_steps_per_second": 47.047,
+      "eval_accuracy": 0.8883870967741936,
+      "eval_loss": 0.0677555724978447,
+      "eval_runtime": 1.3669,
+      "eval_samples_per_second": 2267.902,
+      "eval_steps_per_second": 47.553,
       "step": 954
     },
     {
-      "epoch": 3.14,
-      "learning_rate": 1.2138364779874214e-05,
-      "loss": 0.1675,
+      "epoch": 3.1446540880503147,
+      "grad_norm": 0.5547620058059692,
+      "learning_rate": 1.3011879804332637e-05,
+      "loss": 0.1288,
       "step": 1000
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.917741935483871,
-      "eval_loss": 0.05248909071087837,
-      "eval_runtime": 1.3876,
-      "eval_samples_per_second": 2234.053,
-      "eval_steps_per_second": 46.843,
+      "eval_accuracy": 0.902258064516129,
+      "eval_loss": 0.04977014288306236,
+      "eval_runtime": 1.3624,
+      "eval_samples_per_second": 2275.356,
+      "eval_steps_per_second": 47.709,
       "step": 1272
     },
     {
-      "epoch": 4.72,
-      "learning_rate": 8.207547169811321e-06,
-      "loss": 0.0853,
+      "epoch": 4.716981132075472,
+      "grad_norm": 0.3585783839225769,
+      "learning_rate": 9.517819706498952e-06,
+      "loss": 0.076,
       "step": 1500
     },
     {
       "epoch": 5.0,
-      "eval_accuracy": 0.9274193548387096,
-      "eval_loss": 0.044280897825956345,
-      "eval_runtime": 1.3907,
-      "eval_samples_per_second": 2229.067,
-      "eval_steps_per_second": 46.739,
+      "eval_accuracy": 0.9167741935483871,
+      "eval_loss": 0.039991483092308044,
+      "eval_runtime": 1.3719,
+      "eval_samples_per_second": 2259.57,
+      "eval_steps_per_second": 47.378,
       "step": 1590
     },
     {
       "epoch": 6.0,
-      "eval_accuracy": 0.9293548387096774,
-      "eval_loss": 0.03997669741511345,
-      "eval_runtime": 1.3903,
-      "eval_samples_per_second": 2229.747,
-      "eval_steps_per_second": 46.753,
+      "eval_accuracy": 0.9232258064516129,
+      "eval_loss": 0.034951966255903244,
+      "eval_runtime": 1.3675,
+      "eval_samples_per_second": 2266.904,
+      "eval_steps_per_second": 47.532,
       "step": 1908
     },
     {
-      "epoch": 6.29,
-      "learning_rate": 4.276729559748428e-06,
-      "loss": 0.0651,
+      "epoch": 6.289308176100629,
+      "grad_norm": 0.27236634492874146,
+      "learning_rate": 6.02375960866527e-06,
+      "loss": 0.0585,
       "step": 2000
     },
     {
       "epoch": 7.0,
-      "eval_accuracy": 0.93,
-      "eval_loss": 0.03761696070432663,
-      "eval_runtime": 1.3865,
-      "eval_samples_per_second": 2235.808,
-      "eval_steps_per_second": 46.88,
+      "eval_accuracy": 0.9248387096774193,
+      "eval_loss": 0.032273005694150925,
+      "eval_runtime": 1.3687,
+      "eval_samples_per_second": 2264.964,
+      "eval_steps_per_second": 47.491,
       "step": 2226
     },
     {
-      "epoch": 7.86,
-      "learning_rate": 3.459119496855346e-07,
-      "loss": 0.0581,
+      "epoch": 7.861635220125786,
+      "grad_norm": 0.25957414507865906,
+      "learning_rate": 2.5296995108315863e-06,
+      "loss": 0.0509,
       "step": 2500
     }
   ],
   "logging_steps": 500,
-  "max_steps": 2544,
+  "max_steps": 2862,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 8,
+  "num_train_epochs": 9,
   "save_steps": 500,
-  "total_flos": 719956331598420.0,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 710158307347440.0,
   "train_batch_size": 48,
   "trial_name": null,
   "trial_params": {
-    "alpha": 0.3292104451728688,
-    "num_train_epochs": 8,
-    "temperature": 2
+    "alpha": 0.9970515290608379,
+    "num_train_epochs": 9,
+    "temperature": 4
   }
 }

run-15/checkpoint-500/config.json

@@ -326,6 +326,6 @@
   "sinusoidal_pos_embds": false,
   "tie_weights_": true,
   "torch_dtype": "float32",
-  "transformers_version": "4.37.2",
+  "transformers_version": "4.41.1",
   "vocab_size": 30522
 }

run-15/checkpoint-500/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3c80e66704c8a3c5ffe98c89981aefd5cc485ca7f96b30d8a4b742b213cac452
+oid sha256:c523e1752fda46381be0344a31afc01362450b48bbf8705a2c2bdc36cbf5d0d2
 size 268290900

run-15/checkpoint-500/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0a633eb30b6ab7d592fb74124faa6a15b2c8a08e927bd9699d33aafe9bb5859b
+oid sha256:d08d8a9628761ec315d94e75629b6b7ab53f4005f41e207ed0e3588ae544ab03
 size 536643898

run-15/checkpoint-500/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f8174c1d6ac2627de77161655e4e20ef37d9f2235e2e7c4adc0da0d4e0e14b6d
+oid sha256:b9fb2a0dfb1b0ccef3590fe01e9bd16b6db86f247cc9c7e77290c217a53bac20
 size 14244

run-15/checkpoint-500/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:90dc4637e972cc69c745eebddd8a7560dca27d2318df3e23f8e145abbf236536
+oid sha256:98041bd7cae455426e290a1a0ee683bd5dd30893f7451fec3a464ae8995b17e4
 size 1064

run-15/checkpoint-500/tokenizer.json

@@ -1,11 +1,6 @@
 {
   "version": "1.0",
-  "truncation": {
-    "direction": "Right",
-    "max_length": 512,
-    "strategy": "LongestFirst",
-    "stride": 0
-  },
+  "truncation": null,
   "padding": null,
   "added_tokens": [
     {

run-15/checkpoint-500/trainer_state.json

@@ -10,31 +10,44 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.67,
-      "eval_loss": 0.414095014333725,
-      "eval_runtime": 1.3801,
-      "eval_samples_per_second": 2246.155,
-      "eval_steps_per_second": 47.097,
+      "eval_accuracy": 0.62,
+      "eval_loss": 0.249518021941185,
+      "eval_runtime": 1.3697,
+      "eval_samples_per_second": 2263.294,
+      "eval_steps_per_second": 47.456,
       "step": 318
     },
     {
-      "epoch": 1.57,
-      "learning_rate": 1.606918238993711e-05,
-      "loss": 0.6462,
+      "epoch": 1.5723270440251573,
+      "grad_norm": 0.6149903535842896,
+      "learning_rate": 1.650593990216632e-05,
+      "loss": 0.3991,
       "step": 500
     }
   ],
   "logging_steps": 500,
-  "max_steps": 2544,
+  "max_steps": 2862,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 8,
+  "num_train_epochs": 9,
   "save_steps": 500,
-  "total_flos": 200044162482108.0,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 189074629692144.0,
   "train_batch_size": 48,
   "trial_name": null,
   "trial_params": {
-    "alpha": 0.3292104451728688,
-    "num_train_epochs": 8,
-    "temperature": 2
+    "alpha": 0.9970515290608379,
+    "num_train_epochs": 9,
+    "temperature": 4
   }
 }

run-15/checkpoint-500/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:499353ba7b3dd60f794dd75a66eaf71242d58784f4c315a2120e474f2546f46d
-size 4728
+oid sha256:920ec89dc47e401d4efb361d0fd82b37959f7734e045af2043ed258600a2854b
+size 5176

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8f62c7b1cdbc5f04f78a333fc7ad899d41a314edecf47ae9022088c79c529b2e
+oid sha256:920ec89dc47e401d4efb361d0fd82b37959f7734e045af2043ed258600a2854b
 size 5176