Training in progress, step 2000

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c523e1752fda46381be0344a31afc01362450b48bbf8705a2c2bdc36cbf5d0d2
+oid sha256:94dd719ae3e79ab6db163a2c0423be61fef278b4f047585040f2170742065552
 size 268290900

run-15/checkpoint-1000/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:920ec89dc47e401d4efb361d0fd82b37959f7734e045af2043ed258600a2854b
+oid sha256:9b2e73103b9db907e09d098899dc83e3dbf1869ac8fe958d6aec7beb5cf0db0d
 size 5176

run-15/checkpoint-1500/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:920ec89dc47e401d4efb361d0fd82b37959f7734e045af2043ed258600a2854b
+oid sha256:9b2e73103b9db907e09d098899dc83e3dbf1869ac8fe958d6aec7beb5cf0db0d
 size 5176

run-15/checkpoint-2000/trainer_state.json

@@ -12,9 +12,9 @@
       "epoch": 1.0,
       "eval_accuracy": 0.62,
       "eval_loss": 0.249518021941185,
-      "eval_runtime": 1.3697,
-      "eval_samples_per_second": 2263.294,
-      "eval_steps_per_second": 47.456,
+      "eval_runtime": 1.2519,
+      "eval_samples_per_second": 2476.275,
+      "eval_steps_per_second": 51.922,
       "step": 318
     },
     {
@@ -28,18 +28,18 @@
       "epoch": 2.0,
       "eval_accuracy": 0.8422580645161291,
       "eval_loss": 0.11037396639585495,
-      "eval_runtime": 1.3637,
-      "eval_samples_per_second": 2273.273,
-      "eval_steps_per_second": 47.665,
+      "eval_runtime": 1.2525,
+      "eval_samples_per_second": 2475.052,
+      "eval_steps_per_second": 51.896,
       "step": 636
     },
     {
       "epoch": 3.0,
       "eval_accuracy": 0.8883870967741936,
       "eval_loss": 0.0677555724978447,
-      "eval_runtime": 1.3669,
-      "eval_samples_per_second": 2267.902,
-      "eval_steps_per_second": 47.553,
+      "eval_runtime": 1.2545,
+      "eval_samples_per_second": 2471.149,
+      "eval_steps_per_second": 51.814,
       "step": 954
     },
     {
@@ -53,9 +53,9 @@
       "epoch": 4.0,
       "eval_accuracy": 0.902258064516129,
       "eval_loss": 0.04977014288306236,
-      "eval_runtime": 1.3624,
-      "eval_samples_per_second": 2275.356,
-      "eval_steps_per_second": 47.709,
+      "eval_runtime": 1.2559,
+      "eval_samples_per_second": 2468.428,
+      "eval_steps_per_second": 51.757,
       "step": 1272
     },
     {
@@ -69,18 +69,18 @@
       "epoch": 5.0,
       "eval_accuracy": 0.9167741935483871,
       "eval_loss": 0.039991483092308044,
-      "eval_runtime": 1.3719,
-      "eval_samples_per_second": 2259.57,
-      "eval_steps_per_second": 47.378,
+      "eval_runtime": 1.2562,
+      "eval_samples_per_second": 2467.767,
+      "eval_steps_per_second": 51.743,
       "step": 1590
     },
     {
       "epoch": 6.0,
       "eval_accuracy": 0.9232258064516129,
       "eval_loss": 0.034951966255903244,
-      "eval_runtime": 1.3675,
-      "eval_samples_per_second": 2266.904,
-      "eval_steps_per_second": 47.532,
+      "eval_runtime": 1.2571,
+      "eval_samples_per_second": 2466.006,
+      "eval_steps_per_second": 51.707,
       "step": 1908
     },
     {
@@ -108,11 +108,11 @@
       "attributes": {}
     }
   },
-  "total_flos": 579993747211956.0,
+  "total_flos": 603995663965932.0,
   "train_batch_size": 48,
   "trial_name": null,
   "trial_params": {
-    "alpha": 0.9970515290608379,
+    "alpha": 0.9969607810126598,
     "num_train_epochs": 9,
     "temperature": 4
   }

run-15/checkpoint-2000/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:920ec89dc47e401d4efb361d0fd82b37959f7734e045af2043ed258600a2854b
+oid sha256:9b2e73103b9db907e09d098899dc83e3dbf1869ac8fe958d6aec7beb5cf0db0d
 size 5176

run-16/checkpoint-500/trainer_state.json

@@ -10,25 +10,25 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.594516129032258,
-      "eval_loss": 0.20309074223041534,
-      "eval_runtime": 1.4199,
-      "eval_samples_per_second": 2183.305,
-      "eval_steps_per_second": 45.779,
+      "eval_accuracy": 0.6190322580645161,
+      "eval_loss": 0.25124841928482056,
+      "eval_runtime": 1.2599,
+      "eval_samples_per_second": 2460.478,
+      "eval_steps_per_second": 51.591,
       "step": 318
     },
     {
       "epoch": 1.5723270440251573,
-      "grad_norm": 0.5301857590675354,
-      "learning_rate": 1.685534591194969e-05,
-      "loss": 0.3277,
+      "grad_norm": 0.6181610226631165,
+      "learning_rate": 1.606918238993711e-05,
+      "loss": 0.4004,
       "step": 500
     }
   ],
   "logging_steps": 500,
-  "max_steps": 3180,
+  "max_steps": 2544,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 10,
+  "num_train_epochs": 8,
   "save_steps": 500,
   "stateful_callbacks": {
     "TrainerControl": {
@@ -42,12 +42,12 @@
       "attributes": {}
     }
   },
-  "total_flos": 141540333486132.0,
+  "total_flos": 189074629692144.0,
   "train_batch_size": 48,
   "trial_name": null,
   "trial_params": {
-    "alpha": 0.35485465599924254,
-    "num_train_epochs": 10,
-    "temperature": 9
+    "alpha": 0.06427792762358042,
+    "num_train_epochs": 8,
+    "temperature": 4
   }
 }