Training in progress, step 20, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2b38f39f6f3587258153ec187c194bbaf314c9c5424a3a374dec2ec9eb3dd97e
+oid sha256:11fe44fe8c83218998d60f6d45c46ba671810dc9968ccc81b1ee690d225a83e3
 size 25271744

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fd9390f5b5847d92afd3936548f320903171b28c0280066881a17907d5b2adb9
+oid sha256:3591edc03454ccc3f8357bf58af5ed63716f0aa30d9251b6831ba0aa490d74d5
 size 13685516

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ea30508b2168921b3e75386a046978772f5cbc66bbd566c5d99dd61ea40c26f1
+oid sha256:46fd0e0638d103d958d3a11581fa241a723e98e74de4a1c1eca4f2d49a3e1552
 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": 3.5298969745635986,
-  "best_model_checkpoint": "miner_id_24/checkpoint-10",
-  "epoch": 0.0013662135391761731,
+  "best_metric": 3.345660924911499,
+  "best_model_checkpoint": "miner_id_24/checkpoint-20",
+  "epoch": 0.0027324270783523463,
   "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": 11.323,
       "eval_steps_per_second": 5.662,
       "step": 10
+    },
+    {
+      "epoch": 0.0015028348930937905,
+      "grad_norm": 0.4435620903968811,
+      "learning_rate": 0.00019781476007338058,
+      "loss": 3.464,
+      "step": 11
+    },
+    {
+      "epoch": 0.001639456247011408,
+      "grad_norm": 0.43545401096343994,
+      "learning_rate": 0.0001913545457642601,
+      "loss": 3.4121,
+      "step": 12
+    },
+    {
+      "epoch": 0.0017760776009290251,
+      "grad_norm": 0.5440512299537659,
+      "learning_rate": 0.00018090169943749476,
+      "loss": 3.4155,
+      "step": 13
+    },
+    {
+      "epoch": 0.0019126989548466425,
+      "grad_norm": 0.45815685391426086,
+      "learning_rate": 0.00016691306063588583,
+      "loss": 3.4596,
+      "step": 14
+    },
+    {
+      "epoch": 0.0020493203087642597,
+      "grad_norm": 0.5188146829605103,
+      "learning_rate": 0.00015000000000000001,
+      "loss": 3.4601,
+      "step": 15
+    },
+    {
+      "epoch": 0.0020493203087642597,
+      "eval_loss": 3.417468786239624,
+      "eval_runtime": 273.2435,
+      "eval_samples_per_second": 11.279,
+      "eval_steps_per_second": 5.64,
+      "step": 15
+    },
+    {
+      "epoch": 0.002185941662681877,
+      "grad_norm": 0.5342440605163574,
+      "learning_rate": 0.00013090169943749476,
+      "loss": 3.5198,
+      "step": 16
+    },
+    {
+      "epoch": 0.0023225630165994945,
+      "grad_norm": 0.6059970855712891,
+      "learning_rate": 0.00011045284632676536,
+      "loss": 3.4275,
+      "step": 17
+    },
+    {
+      "epoch": 0.002459184370517112,
+      "grad_norm": 0.5621078014373779,
+      "learning_rate": 8.954715367323468e-05,
+      "loss": 3.3777,
+      "step": 18
+    },
+    {
+      "epoch": 0.0025958057244347293,
+      "grad_norm": 0.37497401237487793,
+      "learning_rate": 6.909830056250527e-05,
+      "loss": 3.2164,
+      "step": 19
+    },
+    {
+      "epoch": 0.0027324270783523463,
+      "grad_norm": 0.44177472591400146,
+      "learning_rate": 5.000000000000002e-05,
+      "loss": 3.3615,
+      "step": 20
+    },
+    {
+      "epoch": 0.0027324270783523463,
+      "eval_loss": 3.345660924911499,
+      "eval_runtime": 272.3123,
+      "eval_samples_per_second": 11.318,
+      "eval_steps_per_second": 5.659,
+      "step": 20
     }
   ],
   "logging_steps": 1,
@@ -129,7 +215,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 511583956500480.0,
+  "total_flos": 1010378314088448.0,
   "train_batch_size": 2,
   "trial_name": null,
   "trial_params": null