Training in progress, epoch 2

logs/events.out.tfevents.1709744087.1dcff7455174.7719.14

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7d3c2fadcdd701848e909bd9223c7acbc003a330e04b141b596c147901d06ab5
-size 5315
+oid sha256:0280ba29c1e42d083e174ffe347911425e4168a61f2568e0d49aa4b399703742
+size 5998

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:be5cd8e0fc75da61a495d38c38f82df737c2424a3ce612eba02b7c23cf9c7018
+oid sha256:57db14253ca1fd241e823b6c203cf94037878d1575b293d3aae91cf1bfa2f479
 size 17549312

run-14/checkpoint-192/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f75fcdaea1e823087014d79cfdbf883b6fe18011905c9b48c573833a61f91ee4
+oid sha256:57db14253ca1fd241e823b6c203cf94037878d1575b293d3aae91cf1bfa2f479
 size 17549312

run-14/checkpoint-192/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:04485baced2d3217b08349f96f7dc7694752b311b348836a7520fa27c6f09b0a
+oid sha256:4714a580ab0f1857e441a746ba669d2cc03fa5b2e5d14b1ab049907ace66167b
 size 35122746

run-14/checkpoint-192/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6a1d9d7ce066a6880a8be2b011c0ae942ea983fccbe1a1704ca9e2483ad33279
+oid sha256:c897b2b825ad8895d62e5d9d02241ccb53c178f4fc9fdffcc7e3f0f96348e454
 size 1064

run-14/checkpoint-192/trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.7974559686888454,
+  "best_metric": 0.8003913894324853,
   "best_model_checkpoint": "tiny-bert-sst2-distilled/run-14/checkpoint-192",
   "epoch": 2.0,
   "eval_steps": 500,
@@ -10,55 +10,55 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "grad_norm": 1.392162561416626,
-      "learning_rate": 0.00018327679510936405,
-      "loss": 0.506,
+      "grad_norm": 2.5439341068267822,
+      "learning_rate": 0.00025636357557273684,
+      "loss": 0.5365,
       "step": 96
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.7945205479452054,
-      "eval_f1": 0.8138297872340425,
-      "eval_loss": 0.43322718143463135,
-      "eval_precision": 0.7439222042139384,
-      "eval_recall": 0.898238747553816,
-      "eval_runtime": 28.335,
-      "eval_samples_per_second": 36.068,
-      "eval_steps_per_second": 1.129,
+      "eval_accuracy": 0.7935420743639922,
+      "eval_f1": 0.8087035358114234,
+      "eval_loss": 0.4461754560470581,
+      "eval_precision": 0.7533783783783784,
+      "eval_recall": 0.87279843444227,
+      "eval_runtime": 25.9122,
+      "eval_samples_per_second": 39.441,
+      "eval_steps_per_second": 1.235,
       "step": 96
     },
     {
       "epoch": 2.0,
-      "grad_norm": 2.713252544403076,
-      "learning_rate": 0.00012218453007290937,
-      "loss": 0.4356,
+      "grad_norm": 4.693763732910156,
+      "learning_rate": 0.00022787873384243272,
+      "loss": 0.4548,
       "step": 192
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.7974559686888454,
-      "eval_f1": 0.8169761273209548,
-      "eval_loss": 0.4169149696826935,
-      "eval_precision": 0.7451612903225806,
-      "eval_recall": 0.9041095890410958,
-      "eval_runtime": 28.1818,
-      "eval_samples_per_second": 36.265,
-      "eval_steps_per_second": 1.135,
+      "eval_accuracy": 0.8003913894324853,
+      "eval_f1": 0.8216783216783217,
+      "eval_loss": 0.43167614936828613,
+      "eval_precision": 0.7424960505529226,
+      "eval_recall": 0.9197651663405088,
+      "eval_runtime": 25.6332,
+      "eval_samples_per_second": 39.87,
+      "eval_steps_per_second": 1.248,
       "step": 192
     }
   ],
   "logging_steps": 500,
-  "max_steps": 384,
+  "max_steps": 960,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 4,
+  "num_train_epochs": 10,
   "save_steps": 500,
   "total_flos": 471390394560.0,
   "train_batch_size": 32,
   "trial_name": null,
   "trial_params": {
-    "alpha": 0.7512024699670536,
-    "learning_rate": 0.00024436906014581875,
-    "num_train_epochs": 4,
-    "temperature": 25
+    "alpha": 0.8320855687119669,
+    "learning_rate": 0.0002848484173030409,
+    "num_train_epochs": 10,
+    "temperature": 8
   }
 }

run-14/checkpoint-192/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:79a362a16e304bed4ef4106f9a75a81f97a0fe20c647127af248c0389326eed5
+oid sha256:d1777509d483b05de90fbacc92d98f5cf027460114d71f11ec490e4a5ee41350
 size 4920