Training in progress, step 700, checkpoint

checkpoint-700/adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bf77a7c1fd44dd3b150147cd2fffe878b7fd689deb8bb07f62f9fb4062a131cf
+oid sha256:44c61a817ae9a6a9cf535809a13db1f2264269e44a2e9570e1f315b75e508bfd
 size 9873829

checkpoint-700/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6c020262f5f9f88db32caf56fa334dbb6625730ab2d8d9a994e491853626359c
+oid sha256:f94b1342df8f36cfcaa61fdce4770893583ef5a873cf88fff8d82d463b534bda
 size 42788

checkpoint-700/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7e8e0d13424de41f2d3dd75843c59f2acf4f5c61cdef60b073f4aee97951614f
+oid sha256:979e545a39bdcc0f0a05a7a60b9b5a34760a3718e374e3397600b77338f73824
 size 14244

checkpoint-700/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:00f0a0254b7e61177c287e87b876b738deea627e195cf94456de79f1af81a13d
+oid sha256:6c6b0ac89ff80cd68518526444f1b0be0e4856b16ef94e25d41524d74be98850
 size 1064

checkpoint-700/trainer_state.json

@@ -1,7 +1,7 @@
 {
-  "best_metric": 0.6392325758934021,
-  "best_model_checkpoint": "output/checkpoint-700",
-  "epoch": 0.5615160934523212,
+  "best_metric": 0.6053406596183777,
+  "best_model_checkpoint": "output/checkpoint-500",
+  "epoch": 0.5612909692292273,
   "eval_steps": 50,
   "global_step": 700,
   "is_hyper_param_search": false,
@@ -11,225 +11,225 @@
     {
       "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
     },
     {
       "epoch": 0.52,
-      "learning_rate": 0.00017659157688540645,
-      "loss": 0.6039,
+      "learning_rate": 0.0001767123287671233,
+      "loss": 0.6103,
       "step": 650
     },
     {
       "epoch": 0.52,
-      "eval_accuracy": 0.6985074626865672,
-      "eval_loss": 0.665508508682251,
-      "eval_runtime": 56.4424,
-      "eval_samples_per_second": 5.935,
-      "eval_steps_per_second": 1.488,
+      "eval_accuracy": 0.6746268656716418,
+      "eval_loss": 0.6098934412002563,
+      "eval_runtime": 56.4929,
+      "eval_samples_per_second": 5.93,
+      "eval_steps_per_second": 1.487,
       "step": 650
     },
     {
       "epoch": 0.56,
-      "learning_rate": 0.00016190009794319292,
-      "loss": 0.6321,
+      "learning_rate": 0.00016203522504892366,
+      "loss": 0.6025,
       "step": 700
     },
     {
       "epoch": 0.56,
-      "eval_accuracy": 0.6716417910447762,
-      "eval_loss": 0.6392325758934021,
-      "eval_runtime": 56.437,
-      "eval_samples_per_second": 5.936,
-      "eval_steps_per_second": 1.488,
+      "eval_accuracy": 0.6895522388059702,
+      "eval_loss": 0.6117912530899048,
+      "eval_runtime": 61.1994,
+      "eval_samples_per_second": 5.474,
+      "eval_steps_per_second": 1.373,
       "step": 700
     }
   ],
   "logging_steps": 50,
-  "max_steps": 1246,
+  "max_steps": 1247,
   "num_train_epochs": 1,
   "save_steps": 100,
-  "total_flos": 2.6209416656252928e+17,
+  "total_flos": 2.6281564386484224e+17,
   "trial_name": null,
   "trial_params": null
 }