Training in progress, step 20, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:62853e226946e57240f5d1f94314dbbe37d7b4de66ec9ddb3186360235efc445
+oid sha256:1c00be323c19379afaa57e4ee047b966c84c68677169025c38dbbace1aefd985
 size 66103768

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4ec173ac65c156c9257b2daaebeee0cd575e546addc86bcf0b21edb6d3781626
+oid sha256:aab82a7b3739ca98654dad952eeaa6f6bdade15066e4d88024222f13ef37659c
 size 33939412

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:72cd4c8a0005b711112f353532d23653255fc2e25e13303c77f47d5ca0a0b7cb
+oid sha256:4769687aee9f547ff5451f7c627b9beab5eac76b248ddebf9af4918dd4f155f5
 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": 3.503246307373047,
-  "best_model_checkpoint": "miner_id_24/checkpoint-10",
-  "epoch": 0.04678362573099415,
+  "best_metric": 2.9236652851104736,
+  "best_model_checkpoint": "miner_id_24/checkpoint-20",
+  "epoch": 0.0935672514619883,
   "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": 5.534,
       "eval_steps_per_second": 2.767,
       "step": 10
+    },
+    {
+      "epoch": 0.05146198830409357,
+      "grad_norm": 4.277874946594238,
+      "learning_rate": 0.00019781476007338058,
+      "loss": 14.2522,
+      "step": 11
+    },
+    {
+      "epoch": 0.056140350877192984,
+      "grad_norm": 3.801171064376831,
+      "learning_rate": 0.0001913545457642601,
+      "loss": 13.4107,
+      "step": 12
+    },
+    {
+      "epoch": 0.0608187134502924,
+      "grad_norm": 4.525282382965088,
+      "learning_rate": 0.00018090169943749476,
+      "loss": 13.6431,
+      "step": 13
+    },
+    {
+      "epoch": 0.06549707602339182,
+      "grad_norm": 4.023133277893066,
+      "learning_rate": 0.00016691306063588583,
+      "loss": 12.3407,
+      "step": 14
+    },
+    {
+      "epoch": 0.07017543859649122,
+      "grad_norm": 5.790004730224609,
+      "learning_rate": 0.00015000000000000001,
+      "loss": 12.456,
+      "step": 15
+    },
+    {
+      "epoch": 0.07017543859649122,
+      "eval_loss": 3.0637786388397217,
+      "eval_runtime": 16.0091,
+      "eval_samples_per_second": 5.622,
+      "eval_steps_per_second": 2.811,
+      "step": 15
+    },
+    {
+      "epoch": 0.07485380116959064,
+      "grad_norm": 6.088311195373535,
+      "learning_rate": 0.00013090169943749476,
+      "loss": 12.2347,
+      "step": 16
+    },
+    {
+      "epoch": 0.07953216374269007,
+      "grad_norm": 5.730970859527588,
+      "learning_rate": 0.00011045284632676536,
+      "loss": 12.0377,
+      "step": 17
+    },
+    {
+      "epoch": 0.08421052631578947,
+      "grad_norm": 5.611118793487549,
+      "learning_rate": 8.954715367323468e-05,
+      "loss": 11.4511,
+      "step": 18
+    },
+    {
+      "epoch": 0.08888888888888889,
+      "grad_norm": 4.9706950187683105,
+      "learning_rate": 6.909830056250527e-05,
+      "loss": 11.446,
+      "step": 19
+    },
+    {
+      "epoch": 0.0935672514619883,
+      "grad_norm": 4.709040641784668,
+      "learning_rate": 5.000000000000002e-05,
+      "loss": 11.6786,
+      "step": 20
+    },
+    {
+      "epoch": 0.0935672514619883,
+      "eval_loss": 2.9236652851104736,
+      "eval_runtime": 16.1604,
+      "eval_samples_per_second": 5.569,
+      "eval_steps_per_second": 2.785,
+      "step": 20
     }
   ],
   "logging_steps": 1,
@@ -129,7 +215,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 1440390402785280.0,
+  "total_flos": 2880780805570560.0,
   "train_batch_size": 2,
   "trial_name": null,
   "trial_params": null