Training in progress, step 20, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ce10e58574d50e06a1662a0ad40583980f2db873d10fa17a181d5b33b9109325
+oid sha256:85f87692dde50c675d641bd22d80eeeaff8845403cabe7eecbcd5b0f9402fdbe
 size 108113968

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c435d41d610426ef99c5b0b10976319fe20e48f7eb7e205922866c302ba5ffc2
+oid sha256:283118d37c87c759e8954796377b1c827c474d39e3d57fb467200fc77b77d790
 size 55549892

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:698bd45eb47251b347466fca533f18c832314c028cb5191e2c2d58bbdd783800
+oid sha256:7a13a8a191f6dd6a178b1ff73b5d7f689cadd10b2afcec900069cb20cda7f689
 size 14244

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bb578e75c11a81e85dda67a691f96ba4793a02960f1409fd3e1511aac873491a
+oid sha256:1d95f3fb1f9df4cdd5f470d1684a242bf1d1940d47b9622802f603a91ffa5bc0
 size 1064

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 1.0585706233978271,
-  "best_model_checkpoint": "miner_id_24/checkpoint-10",
-  "epoch": 0.1282051282051282,
+  "best_metric": 0.8620249032974243,
+  "best_model_checkpoint": "miner_id_24/checkpoint-20",
+  "epoch": 0.2564102564102564,
   "eval_steps": 5,
-  "global_step": 10,
+  "global_step": 20,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -101,6 +101,92 @@
       "eval_samples_per_second": 4.188,
       "eval_steps_per_second": 2.157,
       "step": 10
+    },
+    {
+      "epoch": 0.14102564102564102,
+      "grad_norm": 1.9121575355529785,
+      "learning_rate": 0.00019781476007338058,
+      "loss": 1.0458,
+      "step": 11
+    },
+    {
+      "epoch": 0.15384615384615385,
+      "grad_norm": 1.6511411666870117,
+      "learning_rate": 0.0001913545457642601,
+      "loss": 1.0956,
+      "step": 12
+    },
+    {
+      "epoch": 0.16666666666666666,
+      "grad_norm": 1.3579150438308716,
+      "learning_rate": 0.00018090169943749476,
+      "loss": 0.9845,
+      "step": 13
+    },
+    {
+      "epoch": 0.1794871794871795,
+      "grad_norm": 1.286914587020874,
+      "learning_rate": 0.00016691306063588583,
+      "loss": 0.9154,
+      "step": 14
+    },
+    {
+      "epoch": 0.19230769230769232,
+      "grad_norm": 1.2005118131637573,
+      "learning_rate": 0.00015000000000000001,
+      "loss": 0.999,
+      "step": 15
+    },
+    {
+      "epoch": 0.19230769230769232,
+      "eval_loss": 0.8986483812332153,
+      "eval_runtime": 7.8599,
+      "eval_samples_per_second": 4.199,
+      "eval_steps_per_second": 2.163,
+      "step": 15
+    },
+    {
+      "epoch": 0.20512820512820512,
+      "grad_norm": 0.9289887547492981,
+      "learning_rate": 0.00013090169943749476,
+      "loss": 0.8473,
+      "step": 16
+    },
+    {
+      "epoch": 0.21794871794871795,
+      "grad_norm": 0.8824356198310852,
+      "learning_rate": 0.00011045284632676536,
+      "loss": 0.9723,
+      "step": 17
+    },
+    {
+      "epoch": 0.23076923076923078,
+      "grad_norm": 0.8716819882392883,
+      "learning_rate": 8.954715367323468e-05,
+      "loss": 0.8567,
+      "step": 18
+    },
+    {
+      "epoch": 0.24358974358974358,
+      "grad_norm": 0.875042200088501,
+      "learning_rate": 6.909830056250527e-05,
+      "loss": 0.8004,
+      "step": 19
+    },
+    {
+      "epoch": 0.2564102564102564,
+      "grad_norm": 0.8217772245407104,
+      "learning_rate": 5.000000000000002e-05,
+      "loss": 0.7364,
+      "step": 20
+    },
+    {
+      "epoch": 0.2564102564102564,
+      "eval_loss": 0.8620249032974243,
+      "eval_runtime": 7.6797,
+      "eval_samples_per_second": 4.297,
+      "eval_steps_per_second": 2.214,
+      "step": 20
     }
   ],
   "logging_steps": 1,
@@ -129,7 +215,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 2052393617326080.0,
+  "total_flos": 4104787234652160.0,
   "train_batch_size": 2,
   "trial_name": null,
   "trial_params": null