Training in progress, step 120, checkpoint

checkpoint-120/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fa88b9923d32000ee2b4cdfc9e8d5b21092a959dfc406c3968dcb5b73db420a9
+oid sha256:2dfa054188dae5878f0eecb98d946f0808413058d25d9c28cf47f971f9ebf834
 size 544373707

checkpoint-120/pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0c7ee9f68ca0b9288e0b84b23d77fb474c898c21f60d253a0bbf4b2169d57cba
+oid sha256:3cafb6543dd602e9af54f6677d9014ba0b3e5c563f52768aeccb2fbd2fa44915
 size 272184705

checkpoint-120/trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 7.832551956176758,
+  "best_metric": 7.84664249420166,
   "best_model_checkpoint": "/mmfs1/gscratch/stf/abhinavp/corpus-filtering/outputs/fict-full-lstm-42/checkpoints/checkpoint-120",
   "epoch": 0.96,
   "eval_steps": 10,
@@ -11,169 +11,169 @@
     {
       "epoch": 0.08,
       "learning_rate": 4.600000000000001e-05,
-      "loss": 10.8138,
+      "loss": 10.8142,
       "step": 10
     },
     {
       "epoch": 0.08,
-      "eval_loss": 10.80423355102539,
-      "eval_runtime": 2.8085,
-      "eval_samples_per_second": 356.059,
-      "eval_steps_per_second": 44.507,
+      "eval_loss": 10.805203437805176,
+      "eval_runtime": 2.7628,
+      "eval_samples_per_second": 361.95,
+      "eval_steps_per_second": 45.244,
       "step": 10
     },
     {
       "epoch": 0.16,
       "learning_rate": 4.2e-05,
-      "loss": 10.7946,
+      "loss": 10.7959,
       "step": 20
     },
     {
       "epoch": 0.16,
-      "eval_loss": 10.781017303466797,
-      "eval_runtime": 2.8659,
-      "eval_samples_per_second": 348.927,
-      "eval_steps_per_second": 43.616,
+      "eval_loss": 10.783019065856934,
+      "eval_runtime": 2.9141,
+      "eval_samples_per_second": 343.159,
+      "eval_steps_per_second": 42.895,
       "step": 20
     },
     {
       "epoch": 0.24,
       "learning_rate": 3.8e-05,
-      "loss": 10.7603,
+      "loss": 10.7637,
       "step": 30
     },
     {
       "epoch": 0.24,
-      "eval_loss": 10.720499992370605,
-      "eval_runtime": 2.7648,
-      "eval_samples_per_second": 361.685,
-      "eval_steps_per_second": 45.211,
+      "eval_loss": 10.725115776062012,
+      "eval_runtime": 2.7934,
+      "eval_samples_per_second": 357.988,
+      "eval_steps_per_second": 44.748,
       "step": 30
     },
     {
       "epoch": 0.32,
       "learning_rate": 3.4000000000000007e-05,
-      "loss": 10.6159,
+      "loss": 10.62,
       "step": 40
     },
     {
       "epoch": 0.32,
-      "eval_loss": 10.284521102905273,
-      "eval_runtime": 2.8068,
-      "eval_samples_per_second": 356.277,
-      "eval_steps_per_second": 44.535,
+      "eval_loss": 10.306303977966309,
+      "eval_runtime": 2.7909,
+      "eval_samples_per_second": 358.31,
+      "eval_steps_per_second": 44.789,
       "step": 40
     },
     {
       "epoch": 0.4,
       "learning_rate": 3e-05,
-      "loss": 9.881,
+      "loss": 9.8971,
       "step": 50
     },
     {
       "epoch": 0.4,
-      "eval_loss": 9.241220474243164,
-      "eval_runtime": 2.8265,
-      "eval_samples_per_second": 353.795,
-      "eval_steps_per_second": 44.224,
+      "eval_loss": 9.273224830627441,
+      "eval_runtime": 2.8125,
+      "eval_samples_per_second": 355.55,
+      "eval_steps_per_second": 44.444,
       "step": 50
     },
     {
       "epoch": 0.48,
       "learning_rate": 2.6000000000000002e-05,
-      "loss": 9.0551,
+      "loss": 9.0782,
       "step": 60
     },
     {
       "epoch": 0.48,
-      "eval_loss": 8.63198471069336,
-      "eval_runtime": 2.913,
-      "eval_samples_per_second": 343.289,
-      "eval_steps_per_second": 42.911,
+      "eval_loss": 8.667804718017578,
+      "eval_runtime": 2.8983,
+      "eval_samples_per_second": 345.031,
+      "eval_steps_per_second": 43.129,
       "step": 60
     },
     {
       "epoch": 0.56,
       "learning_rate": 2.2000000000000003e-05,
-      "loss": 8.5603,
+      "loss": 8.5836,
       "step": 70
     },
     {
       "epoch": 0.56,
-      "eval_loss": 8.300483703613281,
-      "eval_runtime": 2.8122,
-      "eval_samples_per_second": 355.598,
-      "eval_steps_per_second": 44.45,
+      "eval_loss": 8.331869125366211,
+      "eval_runtime": 2.7654,
+      "eval_samples_per_second": 361.611,
+      "eval_steps_per_second": 45.201,
       "step": 70
     },
     {
       "epoch": 0.64,
       "learning_rate": 1.8e-05,
-      "loss": 8.3761,
+      "loss": 8.4033,
       "step": 80
     },
     {
       "epoch": 0.64,
-      "eval_loss": 8.105779647827148,
-      "eval_runtime": 2.8076,
-      "eval_samples_per_second": 356.172,
-      "eval_steps_per_second": 44.521,
+      "eval_loss": 8.129607200622559,
+      "eval_runtime": 2.7665,
+      "eval_samples_per_second": 361.474,
+      "eval_steps_per_second": 45.184,
       "step": 80
     },
     {
       "epoch": 0.72,
       "learning_rate": 1.4000000000000001e-05,
-      "loss": 8.2158,
+      "loss": 8.2454,
       "step": 90
     },
     {
       "epoch": 0.72,
-      "eval_loss": 7.982827663421631,
-      "eval_runtime": 2.8051,
-      "eval_samples_per_second": 356.497,
-      "eval_steps_per_second": 44.562,
+      "eval_loss": 8.00214672088623,
+      "eval_runtime": 2.7731,
+      "eval_samples_per_second": 360.605,
+      "eval_steps_per_second": 45.076,
       "step": 90
     },
     {
       "epoch": 0.8,
       "learning_rate": 1e-05,
-      "loss": 8.0922,
+      "loss": 8.1265,
       "step": 100
     },
     {
       "epoch": 0.8,
-      "eval_loss": 7.904599189758301,
-      "eval_runtime": 2.8231,
-      "eval_samples_per_second": 354.225,
-      "eval_steps_per_second": 44.278,
+      "eval_loss": 7.9211931228637695,
+      "eval_runtime": 2.7627,
+      "eval_samples_per_second": 361.96,
+      "eval_steps_per_second": 45.245,
       "step": 100
     },
     {
       "epoch": 0.88,
       "learning_rate": 6e-06,
-      "loss": 8.0326,
+      "loss": 8.0655,
       "step": 110
     },
     {
       "epoch": 0.88,
-      "eval_loss": 7.856319427490234,
-      "eval_runtime": 2.8253,
-      "eval_samples_per_second": 353.939,
-      "eval_steps_per_second": 44.242,
+      "eval_loss": 7.871333122253418,
+      "eval_runtime": 2.7896,
+      "eval_samples_per_second": 358.475,
+      "eval_steps_per_second": 44.809,
       "step": 110
     },
     {
       "epoch": 0.96,
       "learning_rate": 2.0000000000000003e-06,
-      "loss": 8.1135,
+      "loss": 8.1315,
       "step": 120
     },
     {
       "epoch": 0.96,
-      "eval_loss": 7.832551956176758,
-      "eval_runtime": 2.8376,
-      "eval_samples_per_second": 352.411,
-      "eval_steps_per_second": 44.051,
+      "eval_loss": 7.84664249420166,
+      "eval_runtime": 2.9029,
+      "eval_samples_per_second": 344.487,
+      "eval_steps_per_second": 43.061,
       "step": 120
     }
   ],

checkpoint-120/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ca94376b560130659aac7fc1141fac166f9e87d222e6f51cda86dacf6c253994
+oid sha256:baffe53b8f8b613e45dfc80a5db922518a94ed9ef3748800ea2d2e11d33fbc0d
 size 4155