Training in progress, step 2500

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6a6b5a20caef3a60dbe25e51744780de8ee80e367ca8c79d235b3f1e28196b83
+oid sha256:5562e40e2d8c4c5b6d75eaadaf13c35e3662d9300701698371ddeb320098f37a
 size 268290900

run-8/checkpoint-1000/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:692094adc2e16f920342978449186e7f2cd8d6ae9e0526938d74e895a65f10f4
+oid sha256:0d0c54d41e69626eaa1d38e2df7307881acf7d56b84fa8cd48c73da5bab91925
 size 268290900

run-8/checkpoint-1000/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6316013aba05d0de76630b1c30dd5ff13643d9a620b55bd3b0c2e89623c5d747
+oid sha256:bf2af8af45e5e5addf8b25086866db7509d70ff6a036d63377dd3109ba6e22eb
 size 536643898

run-8/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-8/checkpoint-1000/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2d0c5ec14449c9be9b5daeb10d0bc854617fc134f67d60b61b300c34782d3166
-size 4728
+oid sha256:f749d3ae69bc7e3b20b7cd1d8f394bea39a109520f3149e53f47330c0cf61cce
+size 5176

run-8/checkpoint-1500/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0057970acc676abfc730ed9a50cedac13b00203a47d2ea0d541f4d3101f0441c
+oid sha256:42a45feb81fc68209cef779ccddc8347fdc246602423d8ee79e52c992b11b094
 size 268290900

run-8/checkpoint-1500/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8d0a3492313c5d772ed82c8aa3085cd392915b30dfb07a22ee11152bb691f907
+oid sha256:280bd9284ea1c8b42ee189fab53af2cdb70f15b6985d9d1c1c8a692032dc0e8e
 size 536643898

run-8/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-8/checkpoint-1500/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2d0c5ec14449c9be9b5daeb10d0bc854617fc134f67d60b61b300c34782d3166
-size 4728
+oid sha256:f749d3ae69bc7e3b20b7cd1d8f394bea39a109520f3149e53f47330c0cf61cce
+size 5176

run-8/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-8/checkpoint-2000/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a7cd63291a664aa7bf078c050e380aa5250055efbddd6fc3b25873dbf789ba89
+oid sha256:435a53c8e54107bdc0570aa20eaee5791887a3630f7c40ff4f94645fe82c714c
 size 268290900

run-8/checkpoint-2000/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5b9c0db0312cec708615cf97689417a5b5579922029b88f07fb110c4961ff6f5
+oid sha256:8c5bb6be470b5c7d27395588f256ac99dc123e4e5ddbfb2a18b7cd4981e6cadf
 size 536643898

run-8/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-8/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-8/checkpoint-2000/trainer_state.json

@@ -10,80 +10,84 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.6048387096774194,
-      "eval_loss": 0.193294957280159,
-      "eval_runtime": 1.3978,
-      "eval_samples_per_second": 2217.705,
-      "eval_steps_per_second": 46.5,
+      "eval_accuracy": 0.6064516129032258,
+      "eval_loss": 0.2185761034488678,
+      "eval_runtime": 1.2573,
+      "eval_samples_per_second": 2465.677,
+      "eval_steps_per_second": 51.7,
       "step": 318
     },
     {
-      "epoch": 1.57,
+      "epoch": 1.5723270440251573,
+      "grad_norm": 0.5516161918640137,
       "learning_rate": 1.685534591194969e-05,
-      "loss": 0.3109,
+      "loss": 0.3519,
       "step": 500
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.8212903225806452,
-      "eval_loss": 0.09445521980524063,
-      "eval_runtime": 1.3905,
-      "eval_samples_per_second": 2229.493,
-      "eval_steps_per_second": 46.747,
+      "eval_accuracy": 0.8412903225806452,
+      "eval_loss": 0.10163559764623642,
+      "eval_runtime": 1.2573,
+      "eval_samples_per_second": 2465.684,
+      "eval_steps_per_second": 51.7,
       "step": 636
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.8806451612903226,
-      "eval_loss": 0.061732806265354156,
-      "eval_runtime": 1.4039,
-      "eval_samples_per_second": 2208.176,
-      "eval_steps_per_second": 46.3,
+      "eval_accuracy": 0.8858064516129033,
+      "eval_loss": 0.06496834009885788,
+      "eval_runtime": 1.2624,
+      "eval_samples_per_second": 2455.593,
+      "eval_steps_per_second": 51.488,
       "step": 954
     },
     {
-      "epoch": 3.14,
+      "epoch": 3.1446540880503147,
+      "grad_norm": 0.5089602470397949,
       "learning_rate": 1.371069182389937e-05,
-      "loss": 0.1076,
+      "loss": 0.1182,
       "step": 1000
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.9009677419354839,
-      "eval_loss": 0.04701272398233414,
-      "eval_runtime": 1.4228,
-      "eval_samples_per_second": 2178.878,
-      "eval_steps_per_second": 45.686,
+      "eval_accuracy": 0.9003225806451612,
+      "eval_loss": 0.048449017107486725,
+      "eval_runtime": 1.2726,
+      "eval_samples_per_second": 2435.892,
+      "eval_steps_per_second": 51.075,
       "step": 1272
     },
     {
-      "epoch": 4.72,
+      "epoch": 4.716981132075472,
+      "grad_norm": 0.3282025456428528,
       "learning_rate": 1.0566037735849058e-05,
-      "loss": 0.0685,
+      "loss": 0.0719,
       "step": 1500
     },
     {
       "epoch": 5.0,
-      "eval_accuracy": 0.9077419354838709,
-      "eval_loss": 0.03900526091456413,
-      "eval_runtime": 1.5552,
-      "eval_samples_per_second": 1993.305,
-      "eval_steps_per_second": 41.795,
+      "eval_accuracy": 0.9135483870967742,
+      "eval_loss": 0.03883416950702667,
+      "eval_runtime": 1.256,
+      "eval_samples_per_second": 2468.115,
+      "eval_steps_per_second": 51.751,
       "step": 1590
     },
     {
       "epoch": 6.0,
-      "eval_accuracy": 0.9161290322580645,
-      "eval_loss": 0.034293096512556076,
-      "eval_runtime": 1.3898,
-      "eval_samples_per_second": 2230.481,
-      "eval_steps_per_second": 46.768,
+      "eval_accuracy": 0.922258064516129,
+      "eval_loss": 0.03377517685294151,
+      "eval_runtime": 1.257,
+      "eval_samples_per_second": 2466.133,
+      "eval_steps_per_second": 51.709,
       "step": 1908
     },
     {
-      "epoch": 6.29,
+      "epoch": 6.289308176100629,
+      "grad_norm": 0.24476511776447296,
       "learning_rate": 7.421383647798742e-06,
-      "loss": 0.0534,
+      "loss": 0.0554,
       "step": 2000
     }
   ],
@@ -92,12 +96,24 @@
   "num_input_tokens_seen": 0,
   "num_train_epochs": 10,
   "save_steps": 500,
-  "total_flos": 766436892710244.0,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 639855343190448.0,
   "train_batch_size": 48,
   "trial_name": null,
   "trial_params": {
-    "alpha": 0.5027632612026453,
+    "alpha": 0.8038743315868176,
     "num_train_epochs": 10,
-    "temperature": 16
+    "temperature": 6
   }
 }

run-8/checkpoint-2000/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2d0c5ec14449c9be9b5daeb10d0bc854617fc134f67d60b61b300c34782d3166
-size 4728
+oid sha256:f749d3ae69bc7e3b20b7cd1d8f394bea39a109520f3149e53f47330c0cf61cce
+size 5176

run-8/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-8/checkpoint-2500/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1c5480e86a14c3a5a2ed23ef1a3c50cc4071345c9709d7ff4fa42a75714fb2d4
+oid sha256:5562e40e2d8c4c5b6d75eaadaf13c35e3662d9300701698371ddeb320098f37a
 size 268290900

run-8/checkpoint-2500/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6267e7147c56b42a2babf11bb31d0773d90f6a6cf848cd12b413e7b310c7e524
+oid sha256:bbd728c822f7cc7ad773da4adc5666e4c902e7852e11797fe28e4c72c37bfafb
 size 536643898

run-8/checkpoint-2500/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:983ad82afbb61099978c5b5a095761d0cc8a5b9889d078c9a1fcb9751b84b6be
+oid sha256:2fe590653dc8c7914203f68b195f5118633e21a9fa30a5fcd6f1518bae980ac9
 size 14244

run-8/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-8/checkpoint-2500/trainer_state.json

@@ -10,95 +10,100 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.6048387096774194,
-      "eval_loss": 0.193294957280159,
-      "eval_runtime": 1.3978,
-      "eval_samples_per_second": 2217.705,
-      "eval_steps_per_second": 46.5,
+      "eval_accuracy": 0.6064516129032258,
+      "eval_loss": 0.2185761034488678,
+      "eval_runtime": 1.2573,
+      "eval_samples_per_second": 2465.677,
+      "eval_steps_per_second": 51.7,
       "step": 318
     },
     {
-      "epoch": 1.57,
+      "epoch": 1.5723270440251573,
+      "grad_norm": 0.5516161918640137,
       "learning_rate": 1.685534591194969e-05,
-      "loss": 0.3109,
+      "loss": 0.3519,
       "step": 500
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.8212903225806452,
-      "eval_loss": 0.09445521980524063,
-      "eval_runtime": 1.3905,
-      "eval_samples_per_second": 2229.493,
-      "eval_steps_per_second": 46.747,
+      "eval_accuracy": 0.8412903225806452,
+      "eval_loss": 0.10163559764623642,
+      "eval_runtime": 1.2573,
+      "eval_samples_per_second": 2465.684,
+      "eval_steps_per_second": 51.7,
       "step": 636
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.8806451612903226,
-      "eval_loss": 0.061732806265354156,
-      "eval_runtime": 1.4039,
-      "eval_samples_per_second": 2208.176,
-      "eval_steps_per_second": 46.3,
+      "eval_accuracy": 0.8858064516129033,
+      "eval_loss": 0.06496834009885788,
+      "eval_runtime": 1.2624,
+      "eval_samples_per_second": 2455.593,
+      "eval_steps_per_second": 51.488,
       "step": 954
     },
     {
-      "epoch": 3.14,
+      "epoch": 3.1446540880503147,
+      "grad_norm": 0.5089602470397949,
       "learning_rate": 1.371069182389937e-05,
-      "loss": 0.1076,
+      "loss": 0.1182,
       "step": 1000
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.9009677419354839,
-      "eval_loss": 0.04701272398233414,
-      "eval_runtime": 1.4228,
-      "eval_samples_per_second": 2178.878,
-      "eval_steps_per_second": 45.686,
+      "eval_accuracy": 0.9003225806451612,
+      "eval_loss": 0.048449017107486725,
+      "eval_runtime": 1.2726,
+      "eval_samples_per_second": 2435.892,
+      "eval_steps_per_second": 51.075,
       "step": 1272
     },
     {
-      "epoch": 4.72,
+      "epoch": 4.716981132075472,
+      "grad_norm": 0.3282025456428528,
       "learning_rate": 1.0566037735849058e-05,
-      "loss": 0.0685,
+      "loss": 0.0719,
       "step": 1500
     },
     {
       "epoch": 5.0,
-      "eval_accuracy": 0.9077419354838709,
-      "eval_loss": 0.03900526091456413,
-      "eval_runtime": 1.5552,
-      "eval_samples_per_second": 1993.305,
-      "eval_steps_per_second": 41.795,
+      "eval_accuracy": 0.9135483870967742,
+      "eval_loss": 0.03883416950702667,
+      "eval_runtime": 1.256,
+      "eval_samples_per_second": 2468.115,
+      "eval_steps_per_second": 51.751,
       "step": 1590
     },
     {
       "epoch": 6.0,
-      "eval_accuracy": 0.9161290322580645,
-      "eval_loss": 0.034293096512556076,
-      "eval_runtime": 1.3898,
-      "eval_samples_per_second": 2230.481,
-      "eval_steps_per_second": 46.768,
+      "eval_accuracy": 0.922258064516129,
+      "eval_loss": 0.03377517685294151,
+      "eval_runtime": 1.257,
+      "eval_samples_per_second": 2466.133,
+      "eval_steps_per_second": 51.709,
       "step": 1908
     },
     {
-      "epoch": 6.29,
+      "epoch": 6.289308176100629,
+      "grad_norm": 0.24476511776447296,
       "learning_rate": 7.421383647798742e-06,
-      "loss": 0.0534,
+      "loss": 0.0554,
       "step": 2000
     },
     {
       "epoch": 7.0,
-      "eval_accuracy": 0.9196774193548387,
-      "eval_loss": 0.031023623421788216,
-      "eval_runtime": 1.3938,
-      "eval_samples_per_second": 2224.192,
-      "eval_steps_per_second": 46.636,
+      "eval_accuracy": 0.9251612903225807,
+      "eval_loss": 0.03079158626496792,
+      "eval_runtime": 1.2571,
+      "eval_samples_per_second": 2466.063,
+      "eval_steps_per_second": 51.708,
       "step": 2226
     },
     {
-      "epoch": 7.86,
+      "epoch": 7.861635220125786,
+      "grad_norm": 0.23761987686157227,
       "learning_rate": 4.276729559748428e-06,
-      "loss": 0.0463,
+      "loss": 0.0476,
       "step": 2500
     }
   ],
@@ -107,12 +112,24 @@
   "num_input_tokens_seen": 0,
   "num_train_epochs": 10,
   "save_steps": 500,
-  "total_flos": 896297420957400.0,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 770019903325932.0,
   "train_batch_size": 48,
   "trial_name": null,
   "trial_params": {
-    "alpha": 0.5027632612026453,
+    "alpha": 0.8038743315868176,
     "num_train_epochs": 10,
-    "temperature": 16
+    "temperature": 6
   }
 }

run-8/checkpoint-2500/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2d0c5ec14449c9be9b5daeb10d0bc854617fc134f67d60b61b300c34782d3166
-size 4728
+oid sha256:f749d3ae69bc7e3b20b7cd1d8f394bea39a109520f3149e53f47330c0cf61cce
+size 5176

run-8/checkpoint-3000/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-8/checkpoint-3000/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:dc878a30d57a40d44cd984c907862d63dd6d105aaba4ee03da9f29fecb18e47a
+oid sha256:b9eb2ba79453dadd28b5735006b5f8d9d2e1024a31ed0c05f4a42909a7ecf47b
 size 268290900

run-8/checkpoint-3000/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:66e47e9581cd2db5076160953cb22a9a068c87d6a583250843b7c1077f68a7d8
+oid sha256:bb6b985c8061ed40667d0a4d586326a8bf052442a77a3f4fd412897fc4b0256c
 size 536643898

run-8/checkpoint-3000/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d2079317588b57fb166283f8d4e2d7ad479de871cef67689225914d8182ed55b
+oid sha256:88b20597ae6887093d32d8d4b5c49fb0f4f472ac020d8176e4fb05d171fd6885
 size 14244

run-8/checkpoint-3000/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-8/checkpoint-3000/trainer_state.json

@@ -10,119 +10,125 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.6048387096774194,
-      "eval_loss": 0.193294957280159,
-      "eval_runtime": 1.3978,
-      "eval_samples_per_second": 2217.705,
-      "eval_steps_per_second": 46.5,
+      "eval_accuracy": 0.6064516129032258,
+      "eval_loss": 0.2185761034488678,
+      "eval_runtime": 1.2573,
+      "eval_samples_per_second": 2465.677,
+      "eval_steps_per_second": 51.7,
       "step": 318
     },
     {
-      "epoch": 1.57,
+      "epoch": 1.5723270440251573,
+      "grad_norm": 0.5516161918640137,
       "learning_rate": 1.685534591194969e-05,
-      "loss": 0.3109,
+      "loss": 0.3519,
       "step": 500
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.8212903225806452,
-      "eval_loss": 0.09445521980524063,
-      "eval_runtime": 1.3905,
-      "eval_samples_per_second": 2229.493,
-      "eval_steps_per_second": 46.747,
+      "eval_accuracy": 0.8412903225806452,
+      "eval_loss": 0.10163559764623642,
+      "eval_runtime": 1.2573,
+      "eval_samples_per_second": 2465.684,
+      "eval_steps_per_second": 51.7,
       "step": 636
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.8806451612903226,
-      "eval_loss": 0.061732806265354156,
-      "eval_runtime": 1.4039,
-      "eval_samples_per_second": 2208.176,
-      "eval_steps_per_second": 46.3,
+      "eval_accuracy": 0.8858064516129033,
+      "eval_loss": 0.06496834009885788,
+      "eval_runtime": 1.2624,
+      "eval_samples_per_second": 2455.593,
+      "eval_steps_per_second": 51.488,
       "step": 954
     },
     {
-      "epoch": 3.14,
+      "epoch": 3.1446540880503147,
+      "grad_norm": 0.5089602470397949,
       "learning_rate": 1.371069182389937e-05,
-      "loss": 0.1076,
+      "loss": 0.1182,
       "step": 1000
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.9009677419354839,
-      "eval_loss": 0.04701272398233414,
-      "eval_runtime": 1.4228,
-      "eval_samples_per_second": 2178.878,
-      "eval_steps_per_second": 45.686,
+      "eval_accuracy": 0.9003225806451612,
+      "eval_loss": 0.048449017107486725,
+      "eval_runtime": 1.2726,
+      "eval_samples_per_second": 2435.892,
+      "eval_steps_per_second": 51.075,
       "step": 1272
     },
     {
-      "epoch": 4.72,
+      "epoch": 4.716981132075472,
+      "grad_norm": 0.3282025456428528,
       "learning_rate": 1.0566037735849058e-05,
-      "loss": 0.0685,
+      "loss": 0.0719,
       "step": 1500
     },
     {
       "epoch": 5.0,
-      "eval_accuracy": 0.9077419354838709,
-      "eval_loss": 0.03900526091456413,
-      "eval_runtime": 1.5552,
-      "eval_samples_per_second": 1993.305,
-      "eval_steps_per_second": 41.795,
+      "eval_accuracy": 0.9135483870967742,
+      "eval_loss": 0.03883416950702667,
+      "eval_runtime": 1.256,
+      "eval_samples_per_second": 2468.115,
+      "eval_steps_per_second": 51.751,
       "step": 1590
     },
     {
       "epoch": 6.0,
-      "eval_accuracy": 0.9161290322580645,
-      "eval_loss": 0.034293096512556076,
-      "eval_runtime": 1.3898,
-      "eval_samples_per_second": 2230.481,
-      "eval_steps_per_second": 46.768,
+      "eval_accuracy": 0.922258064516129,
+      "eval_loss": 0.03377517685294151,
+      "eval_runtime": 1.257,
+      "eval_samples_per_second": 2466.133,
+      "eval_steps_per_second": 51.709,
       "step": 1908
     },
     {
-      "epoch": 6.29,
+      "epoch": 6.289308176100629,
+      "grad_norm": 0.24476511776447296,
       "learning_rate": 7.421383647798742e-06,
-      "loss": 0.0534,
+      "loss": 0.0554,
       "step": 2000
     },
     {
       "epoch": 7.0,
-      "eval_accuracy": 0.9196774193548387,
-      "eval_loss": 0.031023623421788216,
-      "eval_runtime": 1.3938,
-      "eval_samples_per_second": 2224.192,
-      "eval_steps_per_second": 46.636,
+      "eval_accuracy": 0.9251612903225807,
+      "eval_loss": 0.03079158626496792,
+      "eval_runtime": 1.2571,
+      "eval_samples_per_second": 2466.063,
+      "eval_steps_per_second": 51.708,
       "step": 2226
     },
     {
-      "epoch": 7.86,
+      "epoch": 7.861635220125786,
+      "grad_norm": 0.23761987686157227,
       "learning_rate": 4.276729559748428e-06,
-      "loss": 0.0463,
+      "loss": 0.0476,
       "step": 2500
     },
     {
       "epoch": 8.0,
-      "eval_accuracy": 0.9248387096774193,
-      "eval_loss": 0.029207222163677216,
-      "eval_runtime": 1.3956,
-      "eval_samples_per_second": 2221.267,
-      "eval_steps_per_second": 46.575,
+      "eval_accuracy": 0.93,
+      "eval_loss": 0.028761114925146103,
+      "eval_runtime": 1.2583,
+      "eval_samples_per_second": 2463.725,
+      "eval_steps_per_second": 51.659,
       "step": 2544
     },
     {
       "epoch": 9.0,
-      "eval_accuracy": 0.925483870967742,
-      "eval_loss": 0.028038062155246735,
-      "eval_runtime": 1.3961,
-      "eval_samples_per_second": 2220.392,
-      "eval_steps_per_second": 46.557,
+      "eval_accuracy": 0.93,
+      "eval_loss": 0.027682578191161156,
+      "eval_runtime": 1.2591,
+      "eval_samples_per_second": 2462.059,
+      "eval_steps_per_second": 51.624,
       "step": 2862
     },
     {
-      "epoch": 9.43,
+      "epoch": 9.433962264150944,
+      "grad_norm": 0.2544229328632355,
       "learning_rate": 1.1320754716981133e-06,
-      "loss": 0.0427,
+      "loss": 0.044,
       "step": 3000
     }
   ],
@@ -131,12 +137,24 @@
   "num_input_tokens_seen": 0,
   "num_train_epochs": 10,
   "save_steps": 500,
-  "total_flos": 1026262752005652.0,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 899602859797908.0,
   "train_batch_size": 48,
   "trial_name": null,
   "trial_params": {
-    "alpha": 0.5027632612026453,
+    "alpha": 0.8038743315868176,
     "num_train_epochs": 10,
-    "temperature": 16
+    "temperature": 6
   }
 }

run-8/checkpoint-3000/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2d0c5ec14449c9be9b5daeb10d0bc854617fc134f67d60b61b300c34782d3166
-size 4728
+oid sha256:f749d3ae69bc7e3b20b7cd1d8f394bea39a109520f3149e53f47330c0cf61cce
+size 5176

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ba6ee41bf2e97af70bb5c7bef1f94ebd5979e03ab5954577cc7dff4f754b42f8
+oid sha256:f749d3ae69bc7e3b20b7cd1d8f394bea39a109520f3149e53f47330c0cf61cce
 size 5176