Training in progress, step 200, checkpoint

checkpoint-200/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:edd2511445784260c3e5bef29ce51e94553038b2dffbc438ff973c8f2c0434f4
+oid sha256:d17c363cd05be312b15770b466c1338fec2b141ac20ae3fe7485fd24b8318d8c
 size 19690328

checkpoint-200/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6d20b7c8796c215c0670734dfe2aac3df948cc8b188b200cb24da97c59993bd9
+oid sha256:0962560544030656e61dc01e1d9860980dbf6790ba53119ffc396e82401a5857
 size 38087098

checkpoint-200/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0b4ea83cd76442aa2683aeffa0e079e8bc8f801b046932119a896d48f6a8c29a
+oid sha256:6b698a3ffed4744d0125b17418b2a3d7497f5d9d4c57f4d5e997a2cb6ffd50f2
 size 14244

checkpoint-200/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:45fddbe9119c68279779a206eb04bf02a5dae4550b02e257ac2ea282d998a374
+oid sha256:f5b43ac6655c6879b737216c959393eb2c7f9aec4594023f8e587c864aeb04e5
 size 1064

checkpoint-200/trainer_state.json

@@ -1,7 +1,7 @@
 {
-  "best_metric": 0.47385331988334656,
-  "best_model_checkpoint": "output/checkpoint-150",
-  "epoch": 0.16422046597557222,
+  "best_metric": 0.4239814877510071,
+  "best_model_checkpoint": "output/checkpoint-50",
+  "epoch": 2.888086642599278,
   "eval_steps": 50,
   "global_step": 200,
   "is_hyper_param_search": false,
@@ -9,77 +9,77 @@
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.0,
+      "epoch": 0.01,
       "learning_rate": 0.0,
-      "loss": 0.8903,
+      "loss": 0.3991,
       "step": 1
     },
     {
-      "epoch": 0.04,
-      "learning_rate": 1.2222222222222224e-05,
-      "loss": 0.7189,
+      "epoch": 0.72,
+      "learning_rate": 1.8902439024390243e-05,
+      "loss": 0.3172,
       "step": 50
     },
     {
-      "epoch": 0.04,
-      "eval_accuracy": 0.8228571428571428,
-      "eval_loss": 0.5415178537368774,
-      "eval_runtime": 28.5756,
-      "eval_samples_per_second": 6.124,
-      "eval_steps_per_second": 1.54,
+      "epoch": 0.72,
+      "eval_accuracy": 0.84,
+      "eval_loss": 0.4239814877510071,
+      "eval_runtime": 28.6347,
+      "eval_samples_per_second": 6.111,
+      "eval_steps_per_second": 1.537,
       "step": 50
     },
     {
-      "epoch": 0.08,
-      "learning_rate": 1.8658536585365855e-05,
-      "loss": 0.5723,
+      "epoch": 1.44,
+      "learning_rate": 1.2804878048780488e-05,
+      "loss": 0.1954,
       "step": 100
     },
     {
-      "epoch": 0.08,
-      "eval_accuracy": 0.8,
-      "eval_loss": 0.49520450830459595,
-      "eval_runtime": 28.516,
-      "eval_samples_per_second": 6.137,
-      "eval_steps_per_second": 1.543,
+      "epoch": 1.44,
+      "eval_accuracy": 0.8514285714285714,
+      "eval_loss": 0.5004297494888306,
+      "eval_runtime": 28.6111,
+      "eval_samples_per_second": 6.117,
+      "eval_steps_per_second": 1.538,
       "step": 100
     },
     {
-      "epoch": 0.12,
-      "learning_rate": 1.5609756097560978e-05,
-      "loss": 0.5122,
+      "epoch": 2.17,
+      "learning_rate": 6.707317073170733e-06,
+      "loss": 0.1457,
       "step": 150
     },
     {
-      "epoch": 0.12,
-      "eval_accuracy": 0.7885714285714286,
-      "eval_loss": 0.47385331988334656,
-      "eval_runtime": 28.2004,
-      "eval_samples_per_second": 6.206,
-      "eval_steps_per_second": 1.56,
+      "epoch": 2.17,
+      "eval_accuracy": 0.8628571428571429,
+      "eval_loss": 0.5306527614593506,
+      "eval_runtime": 28.6321,
+      "eval_samples_per_second": 6.112,
+      "eval_steps_per_second": 1.537,
       "step": 150
     },
     {
-      "epoch": 0.16,
-      "learning_rate": 1.2560975609756098e-05,
-      "loss": 0.4831,
+      "epoch": 2.89,
+      "learning_rate": 6.097560975609757e-07,
+      "loss": 0.1312,
       "step": 200
     },
     {
-      "epoch": 0.16,
-      "eval_accuracy": 0.7942857142857143,
-      "eval_loss": 0.49147841334342957,
-      "eval_runtime": 28.6363,
-      "eval_samples_per_second": 6.111,
-      "eval_steps_per_second": 1.537,
+      "epoch": 2.89,
+      "eval_accuracy": 0.8514285714285714,
+      "eval_loss": 0.6020106077194214,
+      "eval_runtime": 28.6498,
+      "eval_samples_per_second": 6.108,
+      "eval_steps_per_second": 1.536,
       "step": 200
     }
   ],
   "logging_steps": 50,
-  "max_steps": 400,
-  "num_train_epochs": 1,
+  "max_steps": 200,
+  "num_train_epochs": 3,
   "save_steps": 50,
-  "total_flos": 7.399936407834624e+16,
+  "total_flos": 7.64381811965952e+16,
   "trial_name": null,
   "trial_params": null
 }

checkpoint-200/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:090099aa07c672025c60437897a7a19f63d8d7bd4284af2562fc632695431956
+oid sha256:d6e1226be91d4f2bcdcde81b49645b5882b397e67c3f85a88d2c0a49245bf8c5
 size 4600