Training in progress, step 640, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7be70d8683f52dfd6ab8a831dd123d31825f4f46f3d60ff6ff495756b2423892
+oid sha256:ec414aa412d347b020c52d6913febda691ffa58147d1fc4ec8933d4bbd5792cc
 size 109009752

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7cef13c342042cdffc93928dccbe66ef17dc59173c61c71746512d665ecab28c
+oid sha256:aa795f21870ee80b58f1c67c34b77a17136213e144d77fd915677c7664bb595e
 size 56088788

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f1da8f1783ffd6f503df78c842d3af807c7c31d93c07e493d90c00923c143bb4
+oid sha256:94499896249a2e4d50fadc9fedec320a01cd48879c1343ddb53ec14a6a37145a
 size 14244

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:74f685c1f5a24a54cae2add4213dad1550d2eddd5d044f686e315a3bdf3dd38b
+oid sha256:a184ad96f2716246808468faf2ca914e9aaf15bc756dd5950c75321082e37ab3
 size 1064

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 2.6299875344193424e-07,
-  "best_model_checkpoint": "miner_id_24/checkpoint-620",
-  "epoch": 4.175084175084175,
+  "best_metric": 2.578611599801661e-07,
+  "best_model_checkpoint": "miner_id_24/checkpoint-640",
+  "epoch": 4.30976430976431,
   "eval_steps": 20,
-  "global_step": 620,
+  "global_step": 640,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -305,6 +305,14 @@
       "eval_samples_per_second": 28.514,
       "eval_steps_per_second": 0.912,
       "step": 620
+    },
+    {
+      "epoch": 4.30976430976431,
+      "eval_loss": 2.578611599801661e-07,
+      "eval_runtime": 4.3645,
+      "eval_samples_per_second": 28.64,
+      "eval_steps_per_second": 0.916,
+      "step": 640
     }
   ],
   "logging_steps": 100,
@@ -333,7 +341,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 2.6180504663359488e+17,
+  "total_flos": 2.7025377746485248e+17,
   "train_batch_size": 8,
   "trial_name": null,
   "trial_params": null