Training in progress, step 20, checkpoint

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a4a9363a2a673c3fe1089907dfeb0e8180df9ee7402d1e8ff77d6dd7b4aa5650
+oid sha256:d6a34638443619a0074ff65dd497b16f574e80c9fdb420e3c4296436112df978
 size 4644026

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ed0777ad4342578dcc46ca5f5a6d5fcf48dd8094a61fe1c87b41cb2f83ff1c6a
+oid sha256:c5dd73064a7e6156867bfe7cc42d099f9b3ca8a81e8092e75ee1ad1a149c9681
 size 14244

last-checkpoint/trainer_state.json

@@ -1,7 +1,7 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 0.0001245204018895971,
+  "epoch": 6.226020094479855e-05,
   "eval_steps": 500,
   "global_step": 20,
   "is_hyper_param_search": false,
@@ -12,7 +12,7 @@
       "epoch": 0.0,
       "grad_norm": NaN,
       "learning_rate": 9.999999999999999e-06,
-      "loss": 4.7178,
+      "loss": 4.7404,
       "step": 20
     }
   ],
@@ -21,8 +21,8 @@
   "num_input_tokens_seen": 0,
   "num_train_epochs": 1,
   "save_steps": 20,
-  "total_flos": 297330828518400.0,
-  "train_batch_size": 2,
+  "total_flos": 276350185052160.0,
+  "train_batch_size": 1,
   "trial_name": null,
   "trial_params": null
 }