Training in progress, step 1500

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7202f94c8991033eb55293356a0828fbb1d3ee311f286dcdcf261aea905bbd61
+oid sha256:db248a54a47b48ea41368243092eb5062af8fb7d0ecc43383c0e0ccd2607315e
 size 268290900

run-4/checkpoint-1000/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e52401c0309d60898756909f056329c107ff5fc79ec57c3b64a5414bc2eec21b
+oid sha256:7a8b1c0e9365da92d349b52df59adf912a4e6d053218291ce8fcae42dc5841e5
 size 268290900

run-4/checkpoint-1000/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:705e84a898bbbcdcb6b8adaff44d902329a187124d69e6fdeda5d7f88fd9807c
+oid sha256:b10400ecf48c1edee7eeb87f1d44ed8b8efd23bb43c4e1b5076b54074c5b9965
 size 536643898

run-4/checkpoint-1000/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:045cf4abe065257226fe11ac1a91dfd4c5b953cb2366365ef8bb18b4cf33e734
+oid sha256:9f182501c34e4ea3ebc7617d27edab7e1367582b147e518cd90295ec7f2eaa0f
 size 1064

run-4/checkpoint-1000/trainer_state.json

@@ -10,53 +10,53 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.5803225806451613,
-      "eval_loss": 0.19497348368167877,
-      "eval_runtime": 5.4007,
-      "eval_samples_per_second": 573.995,
-      "eval_steps_per_second": 12.035,
+      "eval_accuracy": 0.5629032258064516,
+      "eval_loss": 0.20365126430988312,
+      "eval_runtime": 5.4275,
+      "eval_samples_per_second": 571.167,
+      "eval_steps_per_second": 11.976,
       "step": 318
     },
     {
       "epoch": 1.57,
-      "learning_rate": 1.606918238993711e-05,
-      "loss": 0.3105,
+      "learning_rate": 1.371069182389937e-05,
+      "loss": 0.3176,
       "step": 500
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.8316129032258065,
-      "eval_loss": 0.09634843468666077,
-      "eval_runtime": 5.5053,
-      "eval_samples_per_second": 563.09,
-      "eval_steps_per_second": 11.807,
+      "eval_accuracy": 0.8109677419354838,
+      "eval_loss": 0.10676911473274231,
+      "eval_runtime": 5.9736,
+      "eval_samples_per_second": 518.95,
+      "eval_steps_per_second": 10.881,
       "step": 636
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.886774193548387,
-      "eval_loss": 0.0653979703783989,
-      "eval_runtime": 5.4805,
-      "eval_samples_per_second": 565.644,
-      "eval_steps_per_second": 11.86,
+      "eval_accuracy": 0.8651612903225806,
+      "eval_loss": 0.07657339423894882,
+      "eval_runtime": 5.4063,
+      "eval_samples_per_second": 573.407,
+      "eval_steps_per_second": 12.023,
       "step": 954
     },
     {
       "epoch": 3.14,
-      "learning_rate": 1.2138364779874214e-05,
-      "loss": 0.1108,
+      "learning_rate": 7.421383647798742e-06,
+      "loss": 0.1224,
       "step": 1000
     }
   ],
   "logging_steps": 500,
-  "max_steps": 2544,
-  "num_train_epochs": 8,
+  "max_steps": 1590,
+  "num_train_epochs": 5,
   "save_steps": 500,
   "total_flos": 259653090321324.0,
   "trial_name": null,
   "trial_params": {
-    "alpha": 0.39337070180584677,
-    "num_train_epochs": 8,
-    "temperature": 19
+    "alpha": 0.6001886417665624,
+    "num_train_epochs": 5,
+    "temperature": 17
   }
 }

run-4/checkpoint-1000/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:be07dd21844eb14a718c16b339dec6dea8a513202bcc77cdb056d37fbe6528d4
+oid sha256:d6fa4caaf6b88215b9a136c2be7bd09e366c4f83e90d93c02937954f31989165
 size 4664

run-4/checkpoint-1500/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b1d58685898192e73dcc4b773d545ac326afe67accbdcfb46d03f160bb73ee84
+oid sha256:db248a54a47b48ea41368243092eb5062af8fb7d0ecc43383c0e0ccd2607315e
 size 268290900

run-4/checkpoint-1500/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:19e6808ad2a3f5a0593c7b857048a2dee7d525c68e7fbc7249e83e263e3441bf
+oid sha256:f6d9ac30ec659ba96498118ba789f90b170074d0e23190cdf6bcde5207a0a833
 size 536643898

run-4/checkpoint-1500/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:eb7b4645e59a5c5cb66f63e8bcb5f006c535fd0ea63db9ae152dd586fd465b28
+oid sha256:91d97f7c1be085852e2ffd2fe2f8a493a080e2b2858e2f3baf642fe214e035a0
 size 1064

run-4/checkpoint-1500/trainer_state.json

@@ -10,68 +10,68 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.5803225806451613,
-      "eval_loss": 0.19497348368167877,
-      "eval_runtime": 5.4007,
-      "eval_samples_per_second": 573.995,
-      "eval_steps_per_second": 12.035,
+      "eval_accuracy": 0.5629032258064516,
+      "eval_loss": 0.20365126430988312,
+      "eval_runtime": 5.4275,
+      "eval_samples_per_second": 571.167,
+      "eval_steps_per_second": 11.976,
       "step": 318
     },
     {
       "epoch": 1.57,
-      "learning_rate": 1.606918238993711e-05,
-      "loss": 0.3105,
+      "learning_rate": 1.371069182389937e-05,
+      "loss": 0.3176,
       "step": 500
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.8316129032258065,
-      "eval_loss": 0.09634843468666077,
-      "eval_runtime": 5.5053,
-      "eval_samples_per_second": 563.09,
-      "eval_steps_per_second": 11.807,
+      "eval_accuracy": 0.8109677419354838,
+      "eval_loss": 0.10676911473274231,
+      "eval_runtime": 5.9736,
+      "eval_samples_per_second": 518.95,
+      "eval_steps_per_second": 10.881,
       "step": 636
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.886774193548387,
-      "eval_loss": 0.0653979703783989,
-      "eval_runtime": 5.4805,
-      "eval_samples_per_second": 565.644,
-      "eval_steps_per_second": 11.86,
+      "eval_accuracy": 0.8651612903225806,
+      "eval_loss": 0.07657339423894882,
+      "eval_runtime": 5.4063,
+      "eval_samples_per_second": 573.407,
+      "eval_steps_per_second": 12.023,
       "step": 954
     },
     {
       "epoch": 3.14,
-      "learning_rate": 1.2138364779874214e-05,
-      "loss": 0.1108,
+      "learning_rate": 7.421383647798742e-06,
+      "loss": 0.1224,
       "step": 1000
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.9003225806451612,
-      "eval_loss": 0.05045487359166145,
-      "eval_runtime": 5.5506,
-      "eval_samples_per_second": 558.493,
-      "eval_steps_per_second": 11.71,
+      "eval_accuracy": 0.89,
+      "eval_loss": 0.06440269201993942,
+      "eval_runtime": 5.4047,
+      "eval_samples_per_second": 573.576,
+      "eval_steps_per_second": 12.027,
       "step": 1272
     },
     {
       "epoch": 4.72,
-      "learning_rate": 8.207547169811321e-06,
-      "loss": 0.0721,
+      "learning_rate": 1.1320754716981133e-06,
+      "loss": 0.088,
       "step": 1500
     }
   ],
   "logging_steps": 500,
-  "max_steps": 2544,
-  "num_train_epochs": 8,
+  "max_steps": 1590,
+  "num_train_epochs": 5,
   "save_steps": 500,
   "total_flos": 389479376069112.0,
   "trial_name": null,
   "trial_params": {
-    "alpha": 0.39337070180584677,
-    "num_train_epochs": 8,
-    "temperature": 19
+    "alpha": 0.6001886417665624,
+    "num_train_epochs": 5,
+    "temperature": 17
   }
 }

run-4/checkpoint-1500/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:be07dd21844eb14a718c16b339dec6dea8a513202bcc77cdb056d37fbe6528d4
+oid sha256:d6fa4caaf6b88215b9a136c2be7bd09e366c4f83e90d93c02937954f31989165
 size 4664

runs/Jan25_17-40-54_c146da53f02f/events.out.tfevents.1706206976.c146da53f02f.3236.5

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:458a503f1839637b48b6b6217f6decac0b6b8c323afbc614ec4365a2f430cf5d
-size 13124
+oid sha256:ecc3b62a37af00de4b4aa047ab0b19e090d23de158f47518d59397d622af135f
+size 14281