Training in progress, step 20, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4ebe11960f2538fc12696b679ba3845d80854b5d660d851063a7b5fb3c764ffa
+oid sha256:4ea04c62f86354903725e778768a676c9e8755773730a2112c47696a6dbdae13
 size 556856304

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3454dde03e965ea018a24038e0ecfc8c0a7451cbf432d5078dd7fd5bed025cf5
+oid sha256:e1e44db9ba9879524905b188daaee170f5b56ecf9052494cda70cb5c944f9ae0
 size 21599060

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:286cbddbe58e94b390d9c43f9eeef9abf68010ab845f692c08c30a9935964055
+oid sha256:6ca669e357aa0e15c87e40a1da0e0ab9476977a64e76ef375c8f5d113bdeafa4
 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": 3.4941697120666504,
-  "best_model_checkpoint": "miner_id_24/checkpoint-10",
-  "epoch": 0.002055287226389888,
+  "best_metric": 1.9039485454559326,
+  "best_model_checkpoint": "miner_id_24/checkpoint-20",
+  "epoch": 0.004110574452779776,
   "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": 8.488,
       "eval_steps_per_second": 4.246,
       "step": 10
+    },
+    {
+      "epoch": 0.002260815949028877,
+      "grad_norm": 19.169452667236328,
+      "learning_rate": 0.00019781476007338058,
+      "loss": 14.3961,
+      "step": 11
+    },
+    {
+      "epoch": 0.0024663446716678655,
+      "grad_norm": 18.04872703552246,
+      "learning_rate": 0.0001913545457642601,
+      "loss": 12.8396,
+      "step": 12
+    },
+    {
+      "epoch": 0.0026718733943068545,
+      "grad_norm": 18.534652709960938,
+      "learning_rate": 0.00018090169943749476,
+      "loss": 12.0344,
+      "step": 13
+    },
+    {
+      "epoch": 0.002877402116945843,
+      "grad_norm": 16.964826583862305,
+      "learning_rate": 0.00016691306063588583,
+      "loss": 11.4905,
+      "step": 14
+    },
+    {
+      "epoch": 0.003082930839584832,
+      "grad_norm": 15.829866409301758,
+      "learning_rate": 0.00015000000000000001,
+      "loss": 9.6201,
+      "step": 15
+    },
+    {
+      "epoch": 0.003082930839584832,
+      "eval_loss": 2.290395736694336,
+      "eval_runtime": 242.9014,
+      "eval_samples_per_second": 8.436,
+      "eval_steps_per_second": 4.22,
+      "step": 15
+    },
+    {
+      "epoch": 0.003288459562223821,
+      "grad_norm": 16.300065994262695,
+      "learning_rate": 0.00013090169943749476,
+      "loss": 9.4337,
+      "step": 16
+    },
+    {
+      "epoch": 0.0034939882848628095,
+      "grad_norm": 13.852517127990723,
+      "learning_rate": 0.00011045284632676536,
+      "loss": 7.5156,
+      "step": 17
+    },
+    {
+      "epoch": 0.0036995170075017985,
+      "grad_norm": 18.403705596923828,
+      "learning_rate": 8.954715367323468e-05,
+      "loss": 8.2848,
+      "step": 18
+    },
+    {
+      "epoch": 0.003905045730140787,
+      "grad_norm": 19.20428466796875,
+      "learning_rate": 6.909830056250527e-05,
+      "loss": 8.4199,
+      "step": 19
+    },
+    {
+      "epoch": 0.004110574452779776,
+      "grad_norm": 18.174694061279297,
+      "learning_rate": 5.000000000000002e-05,
+      "loss": 7.4398,
+      "step": 20
+    },
+    {
+      "epoch": 0.004110574452779776,
+      "eval_loss": 1.9039485454559326,
+      "eval_runtime": 240.9258,
+      "eval_samples_per_second": 8.505,
+      "eval_steps_per_second": 4.254,
+      "step": 20
     }
   ],
   "logging_steps": 1,
@@ -129,7 +215,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 1305898647552000.0,
+  "total_flos": 2611797295104000.0,
   "train_batch_size": 2,
   "trial_name": null,
   "trial_params": null