Training in progress, step 500

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a4ed14a983a063b04b5c0e74f719ee953be8e1b478635360e23a0e81062435df
+oid sha256:6a6b5a20caef3a60dbe25e51744780de8ee80e367ca8c79d235b3f1e28196b83
 size 268290900

run-4/checkpoint-1500/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:417f67bfc777c996b35c8996ff62c52ec34d71e7a766e99077134c8e94a8f85b
+oid sha256:b7e261c4133c5011993c01561d31449f1945278820742b1a10be67f815d8be65
 size 268290900

run-4/checkpoint-1500/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:aa48a0c978c6a7ede2f05dfd7e1a4fdfa87e493dfd4351367eab5fe63f91142a
+oid sha256:a526f6972f40380fe67908c1a158679ea08f29595d4a7e0b940f91789649cca4
 size 536643898

run-4/checkpoint-1500/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:55c8d3ce0734337fc0c187ca5543b4c70ca45d996531f199209b3a0c2a798109
+oid sha256:ad4d7d251acf36e559c362893a1fb310c9f46b20e8a330025a14b6829ce4ab07
 size 1064

run-4/checkpoint-1500/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:09c1d2b3517a81c25cfd3dc4720ca0f5c5cd50449810325441209751db6b15be
+oid sha256:829aeb9a80fa5e890bae01359822ae021bf174b7265d13bb14d18d678935efbc
 size 5176

run-4/checkpoint-2000/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:36e3328b8fb4a8714c24f51f7d4171f2ba6461c4b6d5913393b71021bff698b8
+oid sha256:0c880fb670519c3b8fd3b38da45a1633e5f4534106637fb8a43cac51424b796d
 size 268290900

run-4/checkpoint-2000/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:19456eaaf80e44986281ea730601eeddcfe0e0c962af2e6358cfd3f16de6611c
+oid sha256:6d154f7a2ab3f1fb62596a00e50a64b5436b8321a0325b989c9e2f26ab5c57f2
 size 536643898

run-4/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-4/checkpoint-2000/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:43d45918bfeb622ba899798234ec7646c7a90dd3f5771db086f5b7dee1d5a530
+oid sha256:f38866eaf1d2baeb52a55cb38ece6ee67f3213265b0830dd06c82a9148795bea
 size 1064

run-4/checkpoint-2000/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5d5d2fff8a65a6470ff21893289269130aa490c201d7c1e86a6b03304d6d7dd8
-size 4728
+oid sha256:829aeb9a80fa5e890bae01359822ae021bf174b7265d13bb14d18d678935efbc
+size 5176

run-6/checkpoint-500/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:150019f90e35dcb4b7f49da2bd990bec1dc64463d39dc2e4c01e0cfc43efd14c
+oid sha256:6a6b5a20caef3a60dbe25e51744780de8ee80e367ca8c79d235b3f1e28196b83
 size 268290900

run-6/checkpoint-500/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:036dc8d61a536755c36e04920350c04dda9f8748aacdee7400f4603af435cbd9
+oid sha256:33e2558a9017819f695c86a2d2a95801009d84d8ff9874e7b19f22dd924876eb
 size 536643898

run-6/checkpoint-500/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9e1264523e958cf7990dc5f42d876cc12129475c4603804cf66868aaf25c2c24
+oid sha256:bb78b1e790f49c38375a9a03add3c9373ef72cf0138709ffb96775535ef084c6
 size 1064

run-6/checkpoint-500/trainer_state.json

@@ -10,25 +10,25 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.5832258064516129,
-      "eval_loss": 0.21771539747714996,
-      "eval_runtime": 1.3209,
-      "eval_samples_per_second": 2346.936,
-      "eval_steps_per_second": 49.21,
+      "eval_accuracy": 0.59,
+      "eval_loss": 0.21894077956676483,
+      "eval_runtime": 1.246,
+      "eval_samples_per_second": 2488.003,
+      "eval_steps_per_second": 52.168,
       "step": 318
     },
     {
       "epoch": 1.5723270440251573,
-      "grad_norm": 0.5158940553665161,
-      "learning_rate": 1.371069182389937e-05,
-      "loss": 0.3408,
+      "grad_norm": 0.5261774063110352,
+      "learning_rate": 1.4758909853249476e-05,
+      "loss": 0.346,
       "step": 500
     }
   ],
   "logging_steps": 500,
-  "max_steps": 1590,
+  "max_steps": 1908,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 5,
+  "num_train_epochs": 6,
   "save_steps": 500,
   "stateful_callbacks": {
     "TrainerControl": {
@@ -42,12 +42,12 @@
       "attributes": {}
     }
   },
-  "total_flos": 130072209152340.0,
+  "total_flos": 213076546446120.0,
   "train_batch_size": 48,
   "trial_name": null,
   "trial_params": {
-    "alpha": 0.21209041164928688,
-    "num_train_epochs": 5,
-    "temperature": 8
+    "alpha": 0.2868213510898018,
+    "num_train_epochs": 6,
+    "temperature": 7
   }
 }

run-6/checkpoint-500/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:03960be7937998204e7d886e253e2b4e234ef43d75169693d208fdea3e6c7062
+oid sha256:ba6ee41bf2e97af70bb5c7bef1f94ebd5979e03ab5954577cc7dff4f754b42f8
 size 5176

run-8/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-8/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-8/checkpoint-1000/trainer_state.json

@@ -10,41 +10,43 @@
   "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
     }
   ],
@@ -53,12 +55,24 @@
   "num_input_tokens_seen": 0,
   "num_train_epochs": 10,
   "save_steps": 500,
-  "total_flos": 507156838101732.0,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 379805351171904.0,
   "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-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-8/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-8/checkpoint-1500/trainer_state.json

@@ -10,56 +10,59 @@
   "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
     }
   ],
@@ -68,12 +71,24 @@
   "num_input_tokens_seen": 0,
   "num_train_epochs": 10,
   "save_steps": 500,
-  "total_flos": 637476008310120.0,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 510232955961624.0,
   "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-500/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b13871d74383d86435fd8abfc1e4e91aec6f513610429f304fa77dc682847ca0
+oid sha256:52acb0f10dbc4d7d9fc1bdc8234f938bf09a424cb2ae024de0d4b496ce371435
 size 268290900

run-8/checkpoint-500/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:079cfbdc10568e390a3e5da7c104fa6b167cafb7d9204a935437f3f49c4625be
+oid sha256:ede2108aa21755bea146deda4c9c419892c1cd9d89a6a7f459b3be1057346deb
 size 536643898

run-8/checkpoint-500/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9e1264523e958cf7990dc5f42d876cc12129475c4603804cf66868aaf25c2c24
+oid sha256:04366f62f8f88f5a8265df59adb051b320463277845db80e7fa43f13110c18c9
 size 1064

run-8/checkpoint-500/trainer_state.json

@@ -10,25 +10,25 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.6590322580645162,
-      "eval_loss": 0.4307226538658142,
-      "eval_runtime": 1.3268,
-      "eval_samples_per_second": 2336.53,
-      "eval_steps_per_second": 48.992,
+      "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": 1.5723270440251573,
-      "grad_norm": 0.942814290523529,
-      "learning_rate": 1.371069182389937e-05,
-      "loss": 0.6582,
+      "grad_norm": 0.5516161918640137,
+      "learning_rate": 1.685534591194969e-05,
+      "loss": 0.3519,
       "step": 500
     }
   ],
   "logging_steps": 500,
-  "max_steps": 1590,
+  "max_steps": 3180,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 5,
+  "num_train_epochs": 10,
   "save_steps": 500,
   "stateful_callbacks": {
     "TrainerControl": {
@@ -46,8 +46,8 @@
   "train_batch_size": 48,
   "trial_name": null,
   "trial_params": {
-    "alpha": 0.5348761545096068,
-    "num_train_epochs": 5,
-    "temperature": 2
+    "alpha": 0.8038743315868176,
+    "num_train_epochs": 10,
+    "temperature": 6
   }
 }

run-8/checkpoint-500/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:54f4065fa2b8e5cb22e349ab79a6f78b551068fff808fab27dd1319edd35db41
+oid sha256:f749d3ae69bc7e3b20b7cd1d8f394bea39a109520f3149e53f47330c0cf61cce
 size 5176

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:829aeb9a80fa5e890bae01359822ae021bf174b7265d13bb14d18d678935efbc
+oid sha256:ba6ee41bf2e97af70bb5c7bef1f94ebd5979e03ab5954577cc7dff4f754b42f8
 size 5176