Training in progress, step 20, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:00944857a9af0d557bbe20daad55ad1daed17322ed454812cc7888070f41dd63
+oid sha256:6b68249116728351764382805db7836289acae47e225c5a75421b679b0b94507
 size 80013120

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5609d6e6043d23c25aadac71abfc12a83ab4b2b33c91f9504ca69b6d2e4320ef
+oid sha256:038525db58b3e7db9ef89d982ed5b480bd74032b6cacd8b72857a38e1faf02a6
 size 41119636

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b6d539fe4d67d4d8fda2e72f75d140391a0aaf6f7ce2131b7b9f8701adbd03b4
+oid sha256:4329764fc299dd85a84b7a55661e86fcab9bb1b815145a2c18430ce0fa85af0f
 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": 0.7933573722839355,
-  "best_model_checkpoint": "miner_id_24/checkpoint-10",
-  "epoch": 0.036596523330283626,
+  "best_metric": 0.6794732809066772,
+  "best_model_checkpoint": "miner_id_24/checkpoint-20",
+  "epoch": 0.07319304666056725,
   "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": 7.28,
       "eval_steps_per_second": 3.672,
       "step": 10
+    },
+    {
+      "epoch": 0.04025617566331199,
+      "grad_norm": 3.100653648376465,
+      "learning_rate": 0.00019781476007338058,
+      "loss": 0.7535,
+      "step": 11
+    },
+    {
+      "epoch": 0.04391582799634035,
+      "grad_norm": 4.180665969848633,
+      "learning_rate": 0.0001913545457642601,
+      "loss": 0.9231,
+      "step": 12
+    },
+    {
+      "epoch": 0.04757548032936871,
+      "grad_norm": 1.741687297821045,
+      "learning_rate": 0.00018090169943749476,
+      "loss": 0.7421,
+      "step": 13
+    },
+    {
+      "epoch": 0.05123513266239707,
+      "grad_norm": 0.9916728138923645,
+      "learning_rate": 0.00016691306063588583,
+      "loss": 0.7143,
+      "step": 14
+    },
+    {
+      "epoch": 0.05489478499542543,
+      "grad_norm": 1.542258620262146,
+      "learning_rate": 0.00015000000000000001,
+      "loss": 0.624,
+      "step": 15
+    },
+    {
+      "epoch": 0.05489478499542543,
+      "eval_loss": 0.6847359538078308,
+      "eval_runtime": 16.3455,
+      "eval_samples_per_second": 7.036,
+      "eval_steps_per_second": 3.548,
+      "step": 15
+    },
+    {
+      "epoch": 0.0585544373284538,
+      "grad_norm": 1.3992561101913452,
+      "learning_rate": 0.00013090169943749476,
+      "loss": 0.704,
+      "step": 16
+    },
+    {
+      "epoch": 0.06221408966148216,
+      "grad_norm": 1.114954948425293,
+      "learning_rate": 0.00011045284632676536,
+      "loss": 0.598,
+      "step": 17
+    },
+    {
+      "epoch": 0.06587374199451052,
+      "grad_norm": 0.8004631996154785,
+      "learning_rate": 8.954715367323468e-05,
+      "loss": 0.6342,
+      "step": 18
+    },
+    {
+      "epoch": 0.06953339432753888,
+      "grad_norm": 2.392939329147339,
+      "learning_rate": 6.909830056250527e-05,
+      "loss": 0.9097,
+      "step": 19
+    },
+    {
+      "epoch": 0.07319304666056725,
+      "grad_norm": 1.1565158367156982,
+      "learning_rate": 5.000000000000002e-05,
+      "loss": 0.5315,
+      "step": 20
+    },
+    {
+      "epoch": 0.07319304666056725,
+      "eval_loss": 0.6794732809066772,
+      "eval_runtime": 16.1432,
+      "eval_samples_per_second": 7.124,
+      "eval_steps_per_second": 3.593,
+      "step": 20
     }
   ],
   "logging_steps": 1,
@@ -129,7 +215,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 1647822923366400.0,
+  "total_flos": 3295645846732800.0,
   "train_batch_size": 2,
   "trial_name": null,
   "trial_params": null