Training in progress, step 20, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5d6f0348c4cc369086d5162e8d2c885896de17e719d3ccc3f3dc85332873521f
+oid sha256:e9c0b534542c403f52162c8352fe5a2d0226a270d058eff62c50fa2d83685bb5
 size 80013120

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:97a976a9e9730b420baaaffbf31dec3fa0c8c5dbeea057b38f65ffe6129ced0e
+oid sha256:c58809080169625e0aaf12dcb5cee2fc94e64d32a94abe366253fecdd50c0ef7
 size 41119636

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ea04cc3ac22ce1d0030ee2ec4c6a340f77b0c8932fd68c5cc068a1a0638110d6
+oid sha256:dac6304c3d282c68b3b54b02e10f7ccb87b4af7f30bf7c763231322312d8a778
 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": 0.8111163377761841,
-  "best_model_checkpoint": "miner_id_24/checkpoint-10",
-  "epoch": 0.036596523330283626,
+  "best_metric": 0.6643534302711487,
+  "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": 5.314,
       "eval_steps_per_second": 2.68,
       "step": 10
+    },
+    {
+      "epoch": 0.04025617566331199,
+      "grad_norm": 3.543696880340576,
+      "learning_rate": 0.00019781476007338058,
+      "loss": 0.7904,
+      "step": 11
+    },
+    {
+      "epoch": 0.04391582799634035,
+      "grad_norm": 4.152765274047852,
+      "learning_rate": 0.0001913545457642601,
+      "loss": 0.9458,
+      "step": 12
+    },
+    {
+      "epoch": 0.04757548032936871,
+      "grad_norm": 1.8694536685943604,
+      "learning_rate": 0.00018090169943749476,
+      "loss": 0.7648,
+      "step": 13
+    },
+    {
+      "epoch": 0.05123513266239707,
+      "grad_norm": 0.97343909740448,
+      "learning_rate": 0.00016691306063588583,
+      "loss": 0.717,
+      "step": 14
+    },
+    {
+      "epoch": 0.05489478499542543,
+      "grad_norm": 1.5022720098495483,
+      "learning_rate": 0.00015000000000000001,
+      "loss": 0.6217,
+      "step": 15
+    },
+    {
+      "epoch": 0.05489478499542543,
+      "eval_loss": 0.6803416609764099,
+      "eval_runtime": 21.8205,
+      "eval_samples_per_second": 5.27,
+      "eval_steps_per_second": 2.658,
+      "step": 15
+    },
+    {
+      "epoch": 0.0585544373284538,
+      "grad_norm": 1.3618319034576416,
+      "learning_rate": 0.00013090169943749476,
+      "loss": 0.7226,
+      "step": 16
+    },
+    {
+      "epoch": 0.06221408966148216,
+      "grad_norm": 1.1633155345916748,
+      "learning_rate": 0.00011045284632676536,
+      "loss": 0.6087,
+      "step": 17
+    },
+    {
+      "epoch": 0.06587374199451052,
+      "grad_norm": 0.8884003162384033,
+      "learning_rate": 8.954715367323468e-05,
+      "loss": 0.6443,
+      "step": 18
+    },
+    {
+      "epoch": 0.06953339432753888,
+      "grad_norm": 2.3892009258270264,
+      "learning_rate": 6.909830056250527e-05,
+      "loss": 0.8875,
+      "step": 19
+    },
+    {
+      "epoch": 0.07319304666056725,
+      "grad_norm": 0.9246272444725037,
+      "learning_rate": 5.000000000000002e-05,
+      "loss": 0.536,
+      "step": 20
+    },
+    {
+      "epoch": 0.07319304666056725,
+      "eval_loss": 0.6643534302711487,
+      "eval_runtime": 21.768,
+      "eval_samples_per_second": 5.283,
+      "eval_steps_per_second": 2.664,
+      "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