Training in progress, step 20, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:51b9022f7ee1da440631f6ba894b3c902080f4fb7495e8abbae020b8c4159e69
+oid sha256:dc4334b869811fe02e72806997caf2965c6515b94afe9fc88bb32759efa5e092
 size 125248064

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ce5faba57b0a6b54250e4ee836d279311c837c4af1b31e4d50ed7156c50c928d
+oid sha256:390e30ae50e39c9f218d126c25e349a51bdf12fe637c55b0f8d68ca93ddb98ac
 size 64219860

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ec7dc567c089b9360385dc88ac6ed83e9342db25848fc49dd9956608ce39fa4a
+oid sha256:8be70d832a1bbff101f33c955af2006518b5495f9530a45087a280e66288a93d
 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": 2.2394750118255615,
-  "best_model_checkpoint": "miner_id_24/checkpoint-10",
-  "epoch": 0.009992505620784412,
+  "best_metric": 1.903188705444336,
+  "best_model_checkpoint": "miner_id_24/checkpoint-20",
+  "epoch": 0.019985011241568824,
   "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": 1.307,
       "eval_steps_per_second": 0.654,
       "step": 10
+    },
+    {
+      "epoch": 0.010991756182862853,
+      "grad_norm": 5.933827877044678,
+      "learning_rate": 0.00019781476007338058,
+      "loss": 9.2207,
+      "step": 11
+    },
+    {
+      "epoch": 0.011991006744941295,
+      "grad_norm": 4.955999374389648,
+      "learning_rate": 0.0001913545457642601,
+      "loss": 8.7849,
+      "step": 12
+    },
+    {
+      "epoch": 0.012990257307019736,
+      "grad_norm": 4.303705215454102,
+      "learning_rate": 0.00018090169943749476,
+      "loss": 8.5175,
+      "step": 13
+    },
+    {
+      "epoch": 0.013989507869098177,
+      "grad_norm": 6.266185760498047,
+      "learning_rate": 0.00016691306063588583,
+      "loss": 8.015,
+      "step": 14
+    },
+    {
+      "epoch": 0.014988758431176618,
+      "grad_norm": 7.1753740310668945,
+      "learning_rate": 0.00015000000000000001,
+      "loss": 9.206,
+      "step": 15
+    },
+    {
+      "epoch": 0.014988758431176618,
+      "eval_loss": 2.0024607181549072,
+      "eval_runtime": 320.3497,
+      "eval_samples_per_second": 1.317,
+      "eval_steps_per_second": 0.659,
+      "step": 15
+    },
+    {
+      "epoch": 0.01598800899325506,
+      "grad_norm": 6.993997097015381,
+      "learning_rate": 0.00013090169943749476,
+      "loss": 7.2703,
+      "step": 16
+    },
+    {
+      "epoch": 0.0169872595553335,
+      "grad_norm": 7.98905086517334,
+      "learning_rate": 0.00011045284632676536,
+      "loss": 7.1496,
+      "step": 17
+    },
+    {
+      "epoch": 0.017986510117411942,
+      "grad_norm": 8.894564628601074,
+      "learning_rate": 8.954715367323468e-05,
+      "loss": 8.2454,
+      "step": 18
+    },
+    {
+      "epoch": 0.01898576067949038,
+      "grad_norm": 9.84389877319336,
+      "learning_rate": 6.909830056250527e-05,
+      "loss": 8.3837,
+      "step": 19
+    },
+    {
+      "epoch": 0.019985011241568824,
+      "grad_norm": 5.84263801574707,
+      "learning_rate": 5.000000000000002e-05,
+      "loss": 6.4959,
+      "step": 20
+    },
+    {
+      "epoch": 0.019985011241568824,
+      "eval_loss": 1.903188705444336,
+      "eval_runtime": 327.3109,
+      "eval_samples_per_second": 1.289,
+      "eval_steps_per_second": 0.645,
+      "step": 20
     }
   ],
   "logging_steps": 1,
@@ -129,7 +215,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 3166224305356800.0,
+  "total_flos": 6332448610713600.0,
   "train_batch_size": 2,
   "trial_name": null,
   "trial_params": null