Training in progress, step 600, checkpoint

checkpoint-600/adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:37deb190b9a2785a382c823980f8127500c6d3f73c4b239e74dcec8c77344a37
+oid sha256:6bb2d766da5712d69d7f67a702645fb6f76a8d14e074c203d0107e0dfd6857f4
 size 9873829

checkpoint-600/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:93b611cdef0597ba9d8fde8d6772930a3494fe1237b4c0d1cf65017e36d1d59f
+oid sha256:35cf271be7f7b823a09c09b2e0390be8773396e009e38f19a09df1d697770e94
 size 42788

checkpoint-600/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2e4ecb2e45257c291e8c9af182865750e96e250ade3e3e9209a900c4910cc861
+oid sha256:7dfaaf2ecc252da97469b3907cf4f35201aa796ca5bdc7846d3479196ccc9cfd
 size 14244

checkpoint-600/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ccd0807ef5cb7094d5a1b66afeb889cf22df8e10454978e5c2c49f52f9ff7f05
+oid sha256:94d861f4bee0cdc054261b3593cd099f26e74568f01bbb992a892477a5635023
 size 1064

checkpoint-600/trainer_state.json

@@ -1,7 +1,7 @@
 {
-  "best_metric": 0.6528732180595398,
-  "best_model_checkpoint": "output/checkpoint-200",
-  "epoch": 0.48129950867341825,
+  "best_metric": 0.6053406596183777,
+  "best_model_checkpoint": "output/checkpoint-500",
+  "epoch": 0.4811065450536233,
   "eval_steps": 50,
   "global_step": 600,
   "is_hyper_param_search": false,
@@ -11,195 +11,195 @@
     {
       "epoch": 0.0,
       "learning_rate": 1.3333333333333332e-06,
-      "loss": 0.785,
+      "loss": 0.6133,
       "step": 1
     },
     {
       "epoch": 0.04,
       "learning_rate": 5.9999999999999995e-05,
-      "loss": 0.6547,
+      "loss": 0.61,
       "step": 50
     },
     {
       "epoch": 0.04,
-      "eval_accuracy": 0.6626865671641791,
-      "eval_loss": 0.8300915956497192,
-      "eval_runtime": 61.5689,
-      "eval_samples_per_second": 5.441,
-      "eval_steps_per_second": 1.364,
+      "eval_accuracy": 0.6895522388059702,
+      "eval_loss": 0.6718239784240723,
+      "eval_runtime": 61.4305,
+      "eval_samples_per_second": 5.453,
+      "eval_steps_per_second": 1.367,
       "step": 50
     },
     {
       "epoch": 0.08,
       "learning_rate": 0.00012666666666666666,
-      "loss": 0.6654,
+      "loss": 0.6233,
       "step": 100
     },
     {
       "epoch": 0.08,
-      "eval_accuracy": 0.6895522388059702,
-      "eval_loss": 0.6663276553153992,
-      "eval_runtime": 56.4633,
-      "eval_samples_per_second": 5.933,
-      "eval_steps_per_second": 1.488,
+      "eval_accuracy": 0.6835820895522388,
+      "eval_loss": 0.695776641368866,
+      "eval_runtime": 61.5801,
+      "eval_samples_per_second": 5.44,
+      "eval_steps_per_second": 1.364,
       "step": 100
     },
     {
       "epoch": 0.12,
       "learning_rate": 0.00019333333333333333,
-      "loss": 0.6491,
+      "loss": 0.6187,
       "step": 150
     },
     {
       "epoch": 0.12,
-      "eval_accuracy": 0.6537313432835821,
-      "eval_loss": 0.9172552824020386,
-      "eval_runtime": 56.4705,
-      "eval_samples_per_second": 5.932,
-      "eval_steps_per_second": 1.488,
+      "eval_accuracy": 0.6895522388059702,
+      "eval_loss": 0.6547008156776428,
+      "eval_runtime": 61.4876,
+      "eval_samples_per_second": 5.448,
+      "eval_steps_per_second": 1.366,
       "step": 150
     },
     {
       "epoch": 0.16,
       "learning_rate": 0.00026,
-      "loss": 0.641,
+      "loss": 0.667,
       "step": 200
     },
     {
       "epoch": 0.16,
-      "eval_accuracy": 0.6865671641791045,
-      "eval_loss": 0.6528732180595398,
-      "eval_runtime": 56.4485,
-      "eval_samples_per_second": 5.935,
-      "eval_steps_per_second": 1.488,
+      "eval_accuracy": 0.6686567164179105,
+      "eval_loss": 0.7339211702346802,
+      "eval_runtime": 61.0301,
+      "eval_samples_per_second": 5.489,
+      "eval_steps_per_second": 1.376,
       "step": 200
     },
     {
       "epoch": 0.2,
-      "learning_rate": 0.0002941234084231146,
-      "loss": 0.6521,
+      "learning_rate": 0.0002941291585127201,
+      "loss": 0.6896,
       "step": 250
     },
     {
       "epoch": 0.2,
-      "eval_accuracy": 0.6835820895522388,
-      "eval_loss": 0.8197052478790283,
-      "eval_runtime": 56.4577,
-      "eval_samples_per_second": 5.934,
-      "eval_steps_per_second": 1.488,
+      "eval_accuracy": 0.6388059701492538,
+      "eval_loss": 0.6455020904541016,
+      "eval_runtime": 61.0356,
+      "eval_samples_per_second": 5.489,
+      "eval_steps_per_second": 1.376,
       "step": 250
     },
     {
       "epoch": 0.24,
-      "learning_rate": 0.00027943192948090106,
-      "loss": 0.638,
+      "learning_rate": 0.0002794520547945205,
+      "loss": 0.6732,
       "step": 300
     },
     {
       "epoch": 0.24,
-      "eval_accuracy": 0.6835820895522388,
-      "eval_loss": 0.6961421370506287,
-      "eval_runtime": 56.4506,
-      "eval_samples_per_second": 5.934,
-      "eval_steps_per_second": 1.488,
+      "eval_accuracy": 0.6388059701492538,
+      "eval_loss": 0.8866483569145203,
+      "eval_runtime": 61.5006,
+      "eval_samples_per_second": 5.447,
+      "eval_steps_per_second": 1.366,
       "step": 300
     },
     {
       "epoch": 0.28,
-      "learning_rate": 0.00026474045053868754,
-      "loss": 0.6857,
+      "learning_rate": 0.0002647749510763209,
+      "loss": 0.6255,
       "step": 350
     },
     {
       "epoch": 0.28,
-      "eval_accuracy": 0.6835820895522388,
-      "eval_loss": 0.7876304388046265,
-      "eval_runtime": 56.4651,
-      "eval_samples_per_second": 5.933,
-      "eval_steps_per_second": 1.488,
+      "eval_accuracy": 0.6746268656716418,
+      "eval_loss": 0.6250954866409302,
+      "eval_runtime": 61.4499,
+      "eval_samples_per_second": 5.452,
+      "eval_steps_per_second": 1.367,
       "step": 350
     },
     {
       "epoch": 0.32,
-      "learning_rate": 0.000250048971596474,
-      "loss": 0.6469,
+      "learning_rate": 0.0002500978473581213,
+      "loss": 0.6522,
       "step": 400
     },
     {
       "epoch": 0.32,
-      "eval_accuracy": 0.6626865671641791,
-      "eval_loss": 0.7487305402755737,
-      "eval_runtime": 56.4489,
-      "eval_samples_per_second": 5.935,
-      "eval_steps_per_second": 1.488,
+      "eval_accuracy": 0.6925373134328359,
+      "eval_loss": 0.7217190861701965,
+      "eval_runtime": 61.1569,
+      "eval_samples_per_second": 5.478,
+      "eval_steps_per_second": 1.374,
       "step": 400
     },
     {
       "epoch": 0.36,
-      "learning_rate": 0.00023535749265426052,
-      "loss": 0.6813,
+      "learning_rate": 0.00023542074363992171,
+      "loss": 0.6592,
       "step": 450
     },
     {
       "epoch": 0.36,
-      "eval_accuracy": 0.6626865671641791,
-      "eval_loss": 0.8447557091712952,
-      "eval_runtime": 56.4428,
-      "eval_samples_per_second": 5.935,
-      "eval_steps_per_second": 1.488,
+      "eval_accuracy": 0.6417910447761194,
+      "eval_loss": 0.636791467666626,
+      "eval_runtime": 61.431,
+      "eval_samples_per_second": 5.453,
+      "eval_steps_per_second": 1.367,
       "step": 450
     },
     {
       "epoch": 0.4,
-      "learning_rate": 0.000220666013712047,
-      "loss": 0.645,
+      "learning_rate": 0.00022074363992172208,
+      "loss": 0.624,
       "step": 500
     },
     {
       "epoch": 0.4,
-      "eval_accuracy": 0.6835820895522388,
-      "eval_loss": 0.6624641418457031,
-      "eval_runtime": 56.4531,
-      "eval_samples_per_second": 5.934,
-      "eval_steps_per_second": 1.488,
+      "eval_accuracy": 0.7044776119402985,
+      "eval_loss": 0.6053406596183777,
+      "eval_runtime": 61.4573,
+      "eval_samples_per_second": 5.451,
+      "eval_steps_per_second": 1.367,
       "step": 500
     },
     {
       "epoch": 0.44,
-      "learning_rate": 0.0002059745347698335,
-      "loss": 0.7044,
+      "learning_rate": 0.0002060665362035225,
+      "loss": 0.6098,
       "step": 550
     },
     {
       "epoch": 0.44,
-      "eval_accuracy": 0.7014925373134329,
-      "eval_loss": 0.63779217004776,
-      "eval_runtime": 56.4382,
-      "eval_samples_per_second": 5.936,
-      "eval_steps_per_second": 1.488,
+      "eval_accuracy": 0.6477611940298508,
+      "eval_loss": 0.8857873678207397,
+      "eval_runtime": 61.4645,
+      "eval_samples_per_second": 5.45,
+      "eval_steps_per_second": 1.367,
       "step": 550
     },
     {
       "epoch": 0.48,
-      "learning_rate": 0.00019128305582761997,
-      "loss": 0.6399,
+      "learning_rate": 0.00019138943248532287,
+      "loss": 0.6357,
       "step": 600
     },
     {
       "epoch": 0.48,
-      "eval_accuracy": 0.6716417910447762,
-      "eval_loss": 0.7784742712974548,
-      "eval_runtime": 56.4356,
-      "eval_samples_per_second": 5.936,
-      "eval_steps_per_second": 1.488,
+      "eval_accuracy": 0.7074626865671642,
+      "eval_loss": 0.607388436794281,
+      "eval_runtime": 61.4941,
+      "eval_samples_per_second": 5.448,
+      "eval_steps_per_second": 1.366,
       "step": 600
     }
   ],
   "logging_steps": 50,
-  "max_steps": 1246,
+  "max_steps": 1247,
   "num_train_epochs": 1,
   "save_steps": 100,
-  "total_flos": 2.245386162757632e+17,
+  "total_flos": 2.2519011770830848e+17,
   "trial_name": null,
   "trial_params": null
 }