Training in progress, step 500

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:71e0adabe57431758e6a472d31246e096a12256d0c48319058002b04b0696af1
+oid sha256:1df6a2e095aa6bbb5d360e7f012cbe4a7929871fd4d4183142dd1f5f93bff5b9
 size 268290900

run-0/checkpoint-1000/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2a6bfc0d04fa062aafe14fdad65aae2ea6adae5c8469541e9b7863a1faad3e9a
+oid sha256:c7ee2b76ad26fda27a87269b8487a8982728f6da76312664825a4e1c5c739251
 size 268290900

run-0/checkpoint-1000/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:80c0e2557f9299a203cad2cf29f7a6e92288b2b70263361fd6952fed79531eed
+oid sha256:61d1902e0f419d4bd955d1ea59a4f898f832344652150ef307c6b77d64babc3a
 size 536643898

run-0/checkpoint-1000/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ed9da1cbe846856af516cf037dd225e7f48e2737ef84c42a9bee0753b8c140ef
+oid sha256:9f182501c34e4ea3ebc7617d27edab7e1367582b147e518cd90295ec7f2eaa0f
 size 1064

run-0/checkpoint-1000/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3995b387f2d0291be709cdbf61f96ec0665c9f1f556146e986cb9b22d69b84bd
+oid sha256:6765f5cb88961cbc2387298f0752026d7dd9093834e5e5d36d91152176870a41
 size 5176

run-0/checkpoint-1500/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c12548a0e1d2999b1e89a6451858c1ba19278e932f73f0ec0a958da3b98809ec
+oid sha256:4485ebbdc70232cda43f157c3a0ba5fcb5086e8bea10a25ff4364587e27d9117
 size 268290900

run-0/checkpoint-1500/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e508a0180bc2432169987e3671ecfea4c7f7399055995ca577258a0697775a3d
+oid sha256:f0309665e784927ec40732c3414ab14cf17bfcb720e1ccbaad3f6ef39737f826
 size 536643898

run-0/checkpoint-1500/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ad4d7d251acf36e559c362893a1fb310c9f46b20e8a330025a14b6829ce4ab07
+oid sha256:91d97f7c1be085852e2ffd2fe2f8a493a080e2b2858e2f3baf642fe214e035a0
 size 1064

run-0/checkpoint-1500/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3995b387f2d0291be709cdbf61f96ec0665c9f1f556146e986cb9b22d69b84bd
+oid sha256:6765f5cb88961cbc2387298f0752026d7dd9093834e5e5d36d91152176870a41
 size 5176

run-1/checkpoint-1000/trainer_state.json

@@ -10,50 +10,50 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.577741935483871,
-      "eval_loss": 0.20344002544879913,
-      "eval_runtime": 1.3567,
-      "eval_samples_per_second": 2284.973,
-      "eval_steps_per_second": 47.911,
+      "eval_accuracy": 0.5951612903225807,
+      "eval_loss": 0.20783506333827972,
+      "eval_runtime": 1.3226,
+      "eval_samples_per_second": 2343.937,
+      "eval_steps_per_second": 49.147,
       "step": 318
     },
     {
       "epoch": 1.5723270440251573,
-      "grad_norm": 0.494202196598053,
-      "learning_rate": 1.4758909853249476e-05,
-      "loss": 0.3218,
+      "grad_norm": 0.5312509536743164,
+      "learning_rate": 1.650593990216632e-05,
+      "loss": 0.3341,
       "step": 500
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.8129032258064516,
-      "eval_loss": 0.10371456295251846,
-      "eval_runtime": 1.3471,
-      "eval_samples_per_second": 2301.291,
-      "eval_steps_per_second": 48.253,
+      "eval_accuracy": 0.8338709677419355,
+      "eval_loss": 0.09971419721841812,
+      "eval_runtime": 1.3325,
+      "eval_samples_per_second": 2326.534,
+      "eval_steps_per_second": 48.782,
       "step": 636
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.8709677419354839,
-      "eval_loss": 0.07233396172523499,
-      "eval_runtime": 1.3503,
-      "eval_samples_per_second": 2295.728,
-      "eval_steps_per_second": 48.136,
+      "eval_accuracy": 0.882258064516129,
+      "eval_loss": 0.06562549620866776,
+      "eval_runtime": 1.3417,
+      "eval_samples_per_second": 2310.56,
+      "eval_steps_per_second": 48.447,
       "step": 954
     },
     {
       "epoch": 3.1446540880503147,
-      "grad_norm": 0.4743537902832031,
-      "learning_rate": 9.517819706498952e-06,
-      "loss": 0.1194,
+      "grad_norm": 0.4943838119506836,
+      "learning_rate": 1.3011879804332637e-05,
+      "loss": 0.1156,
       "step": 1000
     }
   ],
   "logging_steps": 500,
-  "max_steps": 1908,
+  "max_steps": 2862,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 6,
+  "num_train_epochs": 9,
   "save_steps": 500,
   "stateful_callbacks": {
     "TrainerControl": {
@@ -71,8 +71,8 @@
   "train_batch_size": 48,
   "trial_name": null,
   "trial_params": {
-    "alpha": 0.6325268278134527,
-    "num_train_epochs": 6,
-    "temperature": 12
+    "alpha": 0.7941624119034844,
+    "num_train_epochs": 9,
+    "temperature": 8
   }
 }

run-1/checkpoint-1500/trainer_state.json

@@ -10,66 +10,66 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.577741935483871,
-      "eval_loss": 0.20344002544879913,
-      "eval_runtime": 1.3567,
-      "eval_samples_per_second": 2284.973,
-      "eval_steps_per_second": 47.911,
+      "eval_accuracy": 0.5951612903225807,
+      "eval_loss": 0.20783506333827972,
+      "eval_runtime": 1.3226,
+      "eval_samples_per_second": 2343.937,
+      "eval_steps_per_second": 49.147,
       "step": 318
     },
     {
       "epoch": 1.5723270440251573,
-      "grad_norm": 0.494202196598053,
-      "learning_rate": 1.4758909853249476e-05,
-      "loss": 0.3218,
+      "grad_norm": 0.5312509536743164,
+      "learning_rate": 1.650593990216632e-05,
+      "loss": 0.3341,
       "step": 500
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.8129032258064516,
-      "eval_loss": 0.10371456295251846,
-      "eval_runtime": 1.3471,
-      "eval_samples_per_second": 2301.291,
-      "eval_steps_per_second": 48.253,
+      "eval_accuracy": 0.8338709677419355,
+      "eval_loss": 0.09971419721841812,
+      "eval_runtime": 1.3325,
+      "eval_samples_per_second": 2326.534,
+      "eval_steps_per_second": 48.782,
       "step": 636
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.8709677419354839,
-      "eval_loss": 0.07233396172523499,
-      "eval_runtime": 1.3503,
-      "eval_samples_per_second": 2295.728,
-      "eval_steps_per_second": 48.136,
+      "eval_accuracy": 0.882258064516129,
+      "eval_loss": 0.06562549620866776,
+      "eval_runtime": 1.3417,
+      "eval_samples_per_second": 2310.56,
+      "eval_steps_per_second": 48.447,
       "step": 954
     },
     {
       "epoch": 3.1446540880503147,
-      "grad_norm": 0.4743537902832031,
-      "learning_rate": 9.517819706498952e-06,
-      "loss": 0.1194,
+      "grad_norm": 0.4943838119506836,
+      "learning_rate": 1.3011879804332637e-05,
+      "loss": 0.1156,
       "step": 1000
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.8848387096774194,
-      "eval_loss": 0.05824179947376251,
-      "eval_runtime": 1.3553,
-      "eval_samples_per_second": 2287.256,
-      "eval_steps_per_second": 47.959,
+      "eval_accuracy": 0.8958064516129032,
+      "eval_loss": 0.04985633119940758,
+      "eval_runtime": 1.3449,
+      "eval_samples_per_second": 2305.032,
+      "eval_steps_per_second": 48.331,
       "step": 1272
     },
     {
       "epoch": 4.716981132075472,
-      "grad_norm": 0.3467690646648407,
-      "learning_rate": 4.276729559748428e-06,
-      "loss": 0.0813,
+      "grad_norm": 0.3221791386604309,
+      "learning_rate": 9.517819706498952e-06,
+      "loss": 0.0725,
       "step": 1500
     }
   ],
   "logging_steps": 500,
-  "max_steps": 1908,
+  "max_steps": 2862,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 6,
+  "num_train_epochs": 9,
   "save_steps": 500,
   "stateful_callbacks": {
     "TrainerControl": {
@@ -87,8 +87,8 @@
   "train_batch_size": 48,
   "trial_name": null,
   "trial_params": {
-    "alpha": 0.6325268278134527,
-    "num_train_epochs": 6,
-    "temperature": 12
+    "alpha": 0.7941624119034844,
+    "num_train_epochs": 9,
+    "temperature": 8
   }
 }

run-1/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-1/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-1/checkpoint-2000/trainer_state.json

@@ -10,94 +10,110 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.6412903225806451,
-      "eval_loss": 0.230123370885849,
-      "eval_runtime": 1.3595,
-      "eval_samples_per_second": 2280.291,
-      "eval_steps_per_second": 47.813,
+      "eval_accuracy": 0.5951612903225807,
+      "eval_loss": 0.20783506333827972,
+      "eval_runtime": 1.3226,
+      "eval_samples_per_second": 2343.937,
+      "eval_steps_per_second": 49.147,
       "step": 318
     },
     {
-      "epoch": 1.57,
-      "learning_rate": 1.685534591194969e-05,
-      "loss": 0.3693,
+      "epoch": 1.5723270440251573,
+      "grad_norm": 0.5312509536743164,
+      "learning_rate": 1.650593990216632e-05,
+      "loss": 0.3341,
       "step": 500
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.8358064516129032,
-      "eval_loss": 0.10448037087917328,
-      "eval_runtime": 1.3653,
-      "eval_samples_per_second": 2270.569,
-      "eval_steps_per_second": 47.609,
+      "eval_accuracy": 0.8338709677419355,
+      "eval_loss": 0.09971419721841812,
+      "eval_runtime": 1.3325,
+      "eval_samples_per_second": 2326.534,
+      "eval_steps_per_second": 48.782,
       "step": 636
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.8851612903225806,
-      "eval_loss": 0.06413500756025314,
-      "eval_runtime": 1.3667,
-      "eval_samples_per_second": 2268.292,
-      "eval_steps_per_second": 47.561,
+      "eval_accuracy": 0.882258064516129,
+      "eval_loss": 0.06562549620866776,
+      "eval_runtime": 1.3417,
+      "eval_samples_per_second": 2310.56,
+      "eval_steps_per_second": 48.447,
       "step": 954
     },
     {
-      "epoch": 3.14,
-      "learning_rate": 1.371069182389937e-05,
-      "loss": 0.1204,
+      "epoch": 3.1446540880503147,
+      "grad_norm": 0.4943838119506836,
+      "learning_rate": 1.3011879804332637e-05,
+      "loss": 0.1156,
       "step": 1000
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.9041935483870968,
-      "eval_loss": 0.04693836718797684,
-      "eval_runtime": 1.3284,
-      "eval_samples_per_second": 2333.641,
-      "eval_steps_per_second": 48.931,
+      "eval_accuracy": 0.8958064516129032,
+      "eval_loss": 0.04985633119940758,
+      "eval_runtime": 1.3449,
+      "eval_samples_per_second": 2305.032,
+      "eval_steps_per_second": 48.331,
       "step": 1272
     },
     {
-      "epoch": 4.72,
-      "learning_rate": 1.0566037735849058e-05,
-      "loss": 0.0726,
+      "epoch": 4.716981132075472,
+      "grad_norm": 0.3221791386604309,
+      "learning_rate": 9.517819706498952e-06,
+      "loss": 0.0725,
       "step": 1500
     },
     {
       "epoch": 5.0,
-      "eval_accuracy": 0.9180645161290323,
-      "eval_loss": 0.03833732381463051,
-      "eval_runtime": 1.2898,
-      "eval_samples_per_second": 2403.423,
-      "eval_steps_per_second": 50.394,
+      "eval_accuracy": 0.9096774193548387,
+      "eval_loss": 0.04061922803521156,
+      "eval_runtime": 1.3564,
+      "eval_samples_per_second": 2285.502,
+      "eval_steps_per_second": 47.922,
       "step": 1590
     },
     {
       "epoch": 6.0,
-      "eval_accuracy": 0.9235483870967742,
-      "eval_loss": 0.03355677053332329,
-      "eval_runtime": 1.2765,
-      "eval_samples_per_second": 2428.541,
-      "eval_steps_per_second": 50.921,
+      "eval_accuracy": 0.9187096774193548,
+      "eval_loss": 0.03560462221503258,
+      "eval_runtime": 1.3515,
+      "eval_samples_per_second": 2293.716,
+      "eval_steps_per_second": 48.094,
       "step": 1908
     },
     {
-      "epoch": 6.29,
-      "learning_rate": 7.421383647798742e-06,
-      "loss": 0.0554,
+      "epoch": 6.289308176100629,
+      "grad_norm": 0.2550632953643799,
+      "learning_rate": 6.02375960866527e-06,
+      "loss": 0.0569,
       "step": 2000
     }
   ],
   "logging_steps": 500,
-  "max_steps": 3180,
+  "max_steps": 2862,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 10,
+  "num_train_epochs": 9,
   "save_steps": 500,
-  "total_flos": 519271419317532.0,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 520991326672152.0,
   "train_batch_size": 48,
   "trial_name": null,
   "trial_params": {
-    "alpha": 0.7207200745295966,
-    "num_train_epochs": 10,
-    "temperature": 5
+    "alpha": 0.7941624119034844,
+    "num_train_epochs": 9,
+    "temperature": 8
   }
 }

run-1/checkpoint-500/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d1992bc8cbd8cfc6c88b1963eedb9aa5c6b4583be00c396662c3f6a358ddb1ad
+oid sha256:1df6a2e095aa6bbb5d360e7f012cbe4a7929871fd4d4183142dd1f5f93bff5b9
 size 268290900

run-1/checkpoint-500/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:569559a6c1c61ada1010013edab7b010346f26be17388bea8473982069d749b5
+oid sha256:f255b5aed36ba4abd626288657ee63762aafb521c202746c931ad4d10b96a61e
 size 536643898

run-1/checkpoint-500/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bb78b1e790f49c38375a9a03add3c9373ef72cf0138709ffb96775535ef084c6
+oid sha256:98041bd7cae455426e290a1a0ee683bd5dd30893f7451fec3a464ae8995b17e4
 size 1064

run-1/checkpoint-500/trainer_state.json

@@ -10,25 +10,25 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.577741935483871,
-      "eval_loss": 0.20344002544879913,
-      "eval_runtime": 1.3567,
-      "eval_samples_per_second": 2284.973,
-      "eval_steps_per_second": 47.911,
+      "eval_accuracy": 0.5951612903225807,
+      "eval_loss": 0.20783506333827972,
+      "eval_runtime": 1.3226,
+      "eval_samples_per_second": 2343.937,
+      "eval_steps_per_second": 49.147,
       "step": 318
     },
     {
       "epoch": 1.5723270440251573,
-      "grad_norm": 0.494202196598053,
-      "learning_rate": 1.4758909853249476e-05,
-      "loss": 0.3218,
+      "grad_norm": 0.5312509536743164,
+      "learning_rate": 1.650593990216632e-05,
+      "loss": 0.3341,
       "step": 500
     }
   ],
   "logging_steps": 500,
-  "max_steps": 1908,
+  "max_steps": 2862,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 6,
+  "num_train_epochs": 9,
   "save_steps": 500,
   "stateful_callbacks": {
     "TrainerControl": {
@@ -46,8 +46,8 @@
   "train_batch_size": 48,
   "trial_name": null,
   "trial_params": {
-    "alpha": 0.6325268278134527,
-    "num_train_epochs": 6,
-    "temperature": 12
+    "alpha": 0.7941624119034844,
+    "num_train_epochs": 9,
+    "temperature": 8
   }
 }

run-1/checkpoint-500/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:720882e08adee18bc186850828f02514ae1f0c4061921a151ad1b71291c1e3c8
+oid sha256:48a44b718cd5e63b306a632a23ed7f69877efbab8daac27533495007e71fadba
 size 5176

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6765f5cb88961cbc2387298f0752026d7dd9093834e5e5d36d91152176870a41
+oid sha256:48a44b718cd5e63b306a632a23ed7f69877efbab8daac27533495007e71fadba
 size 5176