Training in progress, step 2500

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f5c8f21fecf569753096d63ee4655f7a3d6e2a34b376cea7f305ab1a777db3b3
+oid sha256:9cface9244e3cbfa9a7acccbe33edeb0423f00101b0d72637f8f78e168ccdcc7
 size 268290900

run-0/checkpoint-2500/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:25842382366334e6cf7e5c11a1d3e77ba19d5af6bf8badbf73f820da282a2a36
+oid sha256:9cface9244e3cbfa9a7acccbe33edeb0423f00101b0d72637f8f78e168ccdcc7
 size 268290900

run-0/checkpoint-2500/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:53258b94ed7d406b0ec7ff86369a69748aa54086b3577840e183524120e52746
+oid sha256:59879f71a7f43ec8b6bab6934076138e6b296c45d998607376df0cc9c59ca7e4
 size 536643898

run-0/checkpoint-2500/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d116f7be47ce88249a782ce7aafaf5b673a05bfa0ebdf9f14ab4c327c94c6b04
+oid sha256:b7c943a77651a184c6653dd4706c14c9a27d52136e81942a2b361c74acb9f665
 size 1064

run-0/checkpoint-2500/trainer_state.json

@@ -10,107 +10,107 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.5938709677419355,
-      "eval_loss": 0.19997140765190125,
-      "eval_runtime": 5.5473,
-      "eval_samples_per_second": 558.834,
-      "eval_steps_per_second": 11.717,
+      "eval_accuracy": 0.5767741935483871,
+      "eval_loss": 0.18995507061481476,
+      "eval_runtime": 5.3293,
+      "eval_samples_per_second": 581.689,
+      "eval_steps_per_second": 12.197,
       "step": 318
     },
     {
       "epoch": 1.5723270440251573,
-      "grad_norm": 0.5329213738441467,
-      "learning_rate": 1.685534591194969e-05,
-      "loss": 0.3229,
+      "grad_norm": 0.5155988335609436,
+      "learning_rate": 1.650593990216632e-05,
+      "loss": 0.3062,
       "step": 500
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.8238709677419355,
-      "eval_loss": 0.09559512138366699,
-      "eval_runtime": 5.7978,
-      "eval_samples_per_second": 534.683,
-      "eval_steps_per_second": 11.211,
+      "eval_accuracy": 0.8174193548387096,
+      "eval_loss": 0.09330115467309952,
+      "eval_runtime": 5.6232,
+      "eval_samples_per_second": 551.287,
+      "eval_steps_per_second": 11.559,
       "step": 636
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.8812903225806452,
-      "eval_loss": 0.06311405450105667,
-      "eval_runtime": 5.4988,
-      "eval_samples_per_second": 563.759,
-      "eval_steps_per_second": 11.821,
+      "eval_accuracy": 0.8764516129032258,
+      "eval_loss": 0.06318038702011108,
+      "eval_runtime": 5.4039,
+      "eval_samples_per_second": 573.66,
+      "eval_steps_per_second": 12.028,
       "step": 954
     },
     {
       "epoch": 3.1446540880503147,
-      "grad_norm": 0.44504690170288086,
-      "learning_rate": 1.371069182389937e-05,
-      "loss": 0.1111,
+      "grad_norm": 0.42988139390945435,
+      "learning_rate": 1.3011879804332637e-05,
+      "loss": 0.108,
       "step": 1000
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.9006451612903226,
-      "eval_loss": 0.048286207020282745,
-      "eval_runtime": 5.5577,
-      "eval_samples_per_second": 557.787,
-      "eval_steps_per_second": 11.696,
+      "eval_accuracy": 0.8990322580645161,
+      "eval_loss": 0.04927925392985344,
+      "eval_runtime": 5.4126,
+      "eval_samples_per_second": 572.741,
+      "eval_steps_per_second": 12.009,
       "step": 1272
     },
     {
       "epoch": 4.716981132075472,
-      "grad_norm": 0.3021944761276245,
-      "learning_rate": 1.0566037735849058e-05,
+      "grad_norm": 0.297043114900589,
+      "learning_rate": 9.517819706498952e-06,
       "loss": 0.0699,
       "step": 1500
     },
     {
       "epoch": 5.0,
-      "eval_accuracy": 0.9103225806451613,
-      "eval_loss": 0.03919174522161484,
-      "eval_runtime": 5.6898,
-      "eval_samples_per_second": 544.834,
-      "eval_steps_per_second": 11.424,
+      "eval_accuracy": 0.9074193548387097,
+      "eval_loss": 0.040702205151319504,
+      "eval_runtime": 5.5616,
+      "eval_samples_per_second": 557.389,
+      "eval_steps_per_second": 11.687,
       "step": 1590
     },
     {
       "epoch": 6.0,
-      "eval_accuracy": 0.9132258064516129,
-      "eval_loss": 0.03407873958349228,
-      "eval_runtime": 5.5185,
-      "eval_samples_per_second": 561.746,
-      "eval_steps_per_second": 11.779,
+      "eval_accuracy": 0.9103225806451613,
+      "eval_loss": 0.035840049386024475,
+      "eval_runtime": 5.3906,
+      "eval_samples_per_second": 575.073,
+      "eval_steps_per_second": 12.058,
       "step": 1908
     },
     {
       "epoch": 6.289308176100629,
-      "grad_norm": 0.2565140724182129,
-      "learning_rate": 7.421383647798742e-06,
-      "loss": 0.0542,
+      "grad_norm": 0.26027998328208923,
+      "learning_rate": 6.02375960866527e-06,
+      "loss": 0.0552,
       "step": 2000
     },
     {
       "epoch": 7.0,
-      "eval_accuracy": 0.9229032258064516,
-      "eval_loss": 0.03131110221147537,
-      "eval_runtime": 5.6225,
-      "eval_samples_per_second": 551.352,
-      "eval_steps_per_second": 11.561,
+      "eval_accuracy": 0.917741935483871,
+      "eval_loss": 0.03322926536202431,
+      "eval_runtime": 5.4915,
+      "eval_samples_per_second": 564.507,
+      "eval_steps_per_second": 11.836,
       "step": 2226
     },
     {
       "epoch": 7.861635220125786,
-      "grad_norm": 0.27681249380111694,
-      "learning_rate": 4.276729559748428e-06,
-      "loss": 0.0468,
+      "grad_norm": 0.2637036442756653,
+      "learning_rate": 2.5296995108315863e-06,
+      "loss": 0.0486,
       "step": 2500
     }
   ],
   "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,
   "stateful_callbacks": {
     "TrainerControl": {
@@ -128,8 +128,8 @@
   "train_batch_size": 48,
   "trial_name": null,
   "trial_params": {
-    "alpha": 0.2755961419937909,
-    "num_train_epochs": 10,
-    "temperature": 10
+    "alpha": 0.39616967623231003,
+    "num_train_epochs": 9,
+    "temperature": 20
   }
 }

run-0/checkpoint-2500/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b2da5d40335246ac682533d0ca6c52c386b58cb072bae3a30bf87a5e28562733
+oid sha256:d45fdb811c2d25bdfa6f70464c8f8897df3766bc1abc5d6e057a6e0a40f0dc9f
 size 5240

runs/Oct11_12-01-57_f4efc681eaac/events.out.tfevents.1728648577.f4efc681eaac.505.1

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:978c6a0a7aaaf3a08eafe2ceb07256d9260062a5df902a15a2e242fb4d5433fd
-size 15223
+oid sha256:b5bd367c082ef0816355c43647ec87d18e0a8a3563c46c8a8b15ad3c1872f6ef
+size 16080