Training in progress, step 500

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0bd32a034cff8c6216a16572ad2d88a8986c1f2a93a7e1bba4db41998b4e69ad
+oid sha256:8ced52e0685875b8d675e97aaf647b1f9ccce4784eb0c6a6f8c8641ca10b9da5
 size 268290900

run-3/checkpoint-1908/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:73e211ff00f97e27eb7a5041c75ba965ac8e8ea9a32b4294d2473a19b0bb1ecc
+oid sha256:b84c81075cb7840b3478563b2714e10b15d4a63772e63125433f8b2f68d8fc64
 size 268290900

run-3/checkpoint-1908/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bee3e49a8c9805c734951313417254465b1fbf4b14c6887d84149148ca48c956
+oid sha256:17f5c9e94d1580f6d52ad852c30e123b131534d054766d5866412be5184c923b
 size 536643898

run-3/checkpoint-1908/trainer_state.json

@@ -10,68 +10,68 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.5796774193548387,
-      "eval_loss": 0.20986367762088776,
-      "eval_runtime": 5.1673,
-      "eval_samples_per_second": 599.927,
-      "eval_steps_per_second": 12.579,
+      "eval_accuracy": 0.5980645161290322,
+      "eval_loss": 0.2370469570159912,
+      "eval_runtime": 5.499,
+      "eval_samples_per_second": 563.735,
+      "eval_steps_per_second": 11.82,
       "step": 318
     },
     {
       "epoch": 1.5723270440251573,
-      "grad_norm": 0.5460540056228638,
+      "grad_norm": 0.5828151106834412,
       "learning_rate": 1.4758909853249476e-05,
-      "loss": 0.332,
+      "loss": 0.3737,
       "step": 500
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.8129032258064516,
-      "eval_loss": 0.1050410121679306,
-      "eval_runtime": 5.3586,
-      "eval_samples_per_second": 578.505,
-      "eval_steps_per_second": 12.13,
+      "eval_accuracy": 0.8264516129032258,
+      "eval_loss": 0.11328593641519547,
+      "eval_runtime": 5.8606,
+      "eval_samples_per_second": 528.955,
+      "eval_steps_per_second": 11.091,
       "step": 636
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.8709677419354839,
-      "eval_loss": 0.07279336452484131,
-      "eval_runtime": 5.2066,
-      "eval_samples_per_second": 595.399,
-      "eval_steps_per_second": 12.484,
+      "eval_accuracy": 0.8780645161290322,
+      "eval_loss": 0.0755920559167862,
+      "eval_runtime": 5.4689,
+      "eval_samples_per_second": 566.841,
+      "eval_steps_per_second": 11.885,
       "step": 954
     },
     {
       "epoch": 3.1446540880503147,
-      "grad_norm": 0.4723931849002838,
+      "grad_norm": 0.5158044099807739,
       "learning_rate": 9.517819706498952e-06,
-      "loss": 0.1214,
+      "loss": 0.1316,
       "step": 1000
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.892258064516129,
-      "eval_loss": 0.058892734348773956,
-      "eval_runtime": 5.2068,
-      "eval_samples_per_second": 595.377,
-      "eval_steps_per_second": 12.484,
+      "eval_accuracy": 0.8916129032258064,
+      "eval_loss": 0.0598440058529377,
+      "eval_runtime": 5.4778,
+      "eval_samples_per_second": 565.924,
+      "eval_steps_per_second": 11.866,
       "step": 1272
     },
     {
       "epoch": 4.716981132075472,
-      "grad_norm": 0.3437352180480957,
+      "grad_norm": 0.3741317093372345,
       "learning_rate": 4.276729559748428e-06,
-      "loss": 0.0826,
+      "loss": 0.0865,
       "step": 1500
     },
     {
       "epoch": 5.0,
-      "eval_accuracy": 0.8951612903225806,
-      "eval_loss": 0.05154057592153549,
-      "eval_runtime": 5.4583,
-      "eval_samples_per_second": 567.941,
-      "eval_steps_per_second": 11.908,
+      "eval_accuracy": 0.8996774193548387,
+      "eval_loss": 0.05167784169316292,
+      "eval_runtime": 5.8053,
+      "eval_samples_per_second": 533.998,
+      "eval_steps_per_second": 11.197,
       "step": 1590
     }
   ],
@@ -96,8 +96,8 @@
   "train_batch_size": 48,
   "trial_name": null,
   "trial_params": {
-    "alpha": 0.9907493566825466,
+    "alpha": 0.5973659898080927,
     "num_train_epochs": 6,
-    "temperature": 9
+    "temperature": 5
   }
 }

run-3/checkpoint-1908/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a85143060eb3d68597fa3be18a06447136e3eefcb180aa3b82a2a8463f5f692a
+oid sha256:a9bfc1f73d75af662c164817dfdb380c8286c5e24841175e1b2797e416869742
 size 5240

run-4/checkpoint-500/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cffdaf73e0b6a284c8442673d7ba025740d7309ed5dfb3d7ac54c473816a869a
+oid sha256:8ced52e0685875b8d675e97aaf647b1f9ccce4784eb0c6a6f8c8641ca10b9da5
 size 268290900

run-4/checkpoint-500/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fb284d6406695d9c68ff21422fe32da7603f4d1b5919d5a7f09a05ec0ea28dd3
+oid sha256:ff71638289596e8d4bc9e4bc53149603b4613d3f4c04997a6193a147454b1c73
 size 536643898

run-4/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-4/checkpoint-500/trainer_state.json

@@ -10,25 +10,25 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.6325806451612903,
-      "eval_loss": 0.3024460971355438,
-      "eval_runtime": 5.4182,
-      "eval_samples_per_second": 572.151,
-      "eval_steps_per_second": 11.997,
+      "eval_accuracy": 0.6406451612903226,
+      "eval_loss": 0.2967454195022583,
+      "eval_runtime": 5.4866,
+      "eval_samples_per_second": 565.017,
+      "eval_steps_per_second": 11.847,
       "step": 318
     },
     {
       "epoch": 1.5723270440251573,
-      "grad_norm": 0.6661854386329651,
-      "learning_rate": 1.371069182389937e-05,
-      "loss": 0.4696,
+      "grad_norm": 0.6655566096305847,
+      "learning_rate": 1.4758909853249476e-05,
+      "loss": 0.4655,
       "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": {
@@ -46,8 +46,8 @@
   "train_batch_size": 48,
   "trial_name": null,
   "trial_params": {
-    "alpha": 0.41539799624272716,
-    "num_train_epochs": 5,
+    "alpha": 0.6003183502544485,
+    "num_train_epochs": 6,
     "temperature": 3
   }
 }

run-4/checkpoint-500/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:42fcda7114a3cb0f210fff0e0a799f161a86990fe91444d8130cb28445b06005
+oid sha256:39e2230760e0d209a41a97071088ff5c736b18829fcf11d0286149c73a45dbed
 size 5240

runs/Oct12_06-40-39_b76c1be2ae55/events.out.tfevents.1728718423.b76c1be2ae55.1423.5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:db198a5f0e14ccda4309965ca889bf39e9b67c3eab3499a77df07f843e475e4b
+size 14155

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a9bfc1f73d75af662c164817dfdb380c8286c5e24841175e1b2797e416869742
+oid sha256:39e2230760e0d209a41a97071088ff5c736b18829fcf11d0286149c73a45dbed
 size 5240