Training in progress, step 3000, checkpoint

last-checkpoint/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8f47812f8b3a00767b51d45649e9d1a922c140415c54cb3da117207155e040ea
+oid sha256:170bf6c92ae110e8eb5c1a3c0f5e69f21a3de0e3e4a77fd59d30a079ac9d4811
 size 577789320

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f09da17ed80492ce22211b6d6425439ec3c816ab1df397068bdb7c4095422ead
+oid sha256:7cb04e2f0ab69bfcb0d47f3b2af07eb3e8c2cb14295c7e2962418f07502fe274
 size 1155772233

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e0a3df38c17b475ea7e47b6c462cb3838a91cdec2a3f47a69fc35d66481cfa21
+oid sha256:b0bd26a66f18a63513b4644dff9bbce4420bc172da7cc36bbc9e2bbc11796d8c
 size 14244

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:da529ccf5d02ea297cc830d27155ea390510525858813c5d74514201ace90deb
+oid sha256:1f58257cf138830d5c1b5b1b46ab56c0eb866a6a719b4e3b6cce54569429bd96
 size 1064

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 0.35398951172828674,
-  "best_model_checkpoint": "mikhail_panzo/zlm_b128_le4_s8000/checkpoint-2500",
-  "epoch": 4.18848167539267,
+  "best_metric": 0.3458922803401947,
+  "best_model_checkpoint": "mikhail_panzo/zlm_b128_le4_s8000/checkpoint-3000",
+  "epoch": 5.026178010471204,
   "eval_steps": 500,
-  "global_step": 2500,
+  "global_step": 3000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -397,6 +397,84 @@
       "eval_samples_per_second": 31.28,
       "eval_steps_per_second": 3.913,
       "step": 2500
+    },
+    {
+      "epoch": 4.272251308900524,
+      "grad_norm": 2.468984842300415,
+      "learning_rate": 9.451000000000002e-05,
+      "loss": 0.3902,
+      "step": 2550
+    },
+    {
+      "epoch": 4.356020942408377,
+      "grad_norm": 1.2810943126678467,
+      "learning_rate": 9.401e-05,
+      "loss": 0.386,
+      "step": 2600
+    },
+    {
+      "epoch": 4.439790575916231,
+      "grad_norm": 1.6781765222549438,
+      "learning_rate": 9.351e-05,
+      "loss": 0.383,
+      "step": 2650
+    },
+    {
+      "epoch": 4.523560209424084,
+      "grad_norm": 1.617163896560669,
+      "learning_rate": 9.301e-05,
+      "loss": 0.3849,
+      "step": 2700
+    },
+    {
+      "epoch": 4.607329842931938,
+      "grad_norm": 1.4169151782989502,
+      "learning_rate": 9.251000000000001e-05,
+      "loss": 0.3807,
+      "step": 2750
+    },
+    {
+      "epoch": 4.69109947643979,
+      "grad_norm": 1.1944037675857544,
+      "learning_rate": 9.201000000000001e-05,
+      "loss": 0.3838,
+      "step": 2800
+    },
+    {
+      "epoch": 4.774869109947644,
+      "grad_norm": 1.7312718629837036,
+      "learning_rate": 9.151000000000001e-05,
+      "loss": 0.3808,
+      "step": 2850
+    },
+    {
+      "epoch": 4.858638743455497,
+      "grad_norm": 1.357228398323059,
+      "learning_rate": 9.101000000000001e-05,
+      "loss": 0.3832,
+      "step": 2900
+    },
+    {
+      "epoch": 4.942408376963351,
+      "grad_norm": 1.2495553493499756,
+      "learning_rate": 9.051000000000001e-05,
+      "loss": 0.3837,
+      "step": 2950
+    },
+    {
+      "epoch": 5.026178010471204,
+      "grad_norm": 1.3688994646072388,
+      "learning_rate": 9.001e-05,
+      "loss": 0.3802,
+      "step": 3000
+    },
+    {
+      "epoch": 5.026178010471204,
+      "eval_loss": 0.3458922803401947,
+      "eval_runtime": 277.371,
+      "eval_samples_per_second": 30.605,
+      "eval_steps_per_second": 3.829,
+      "step": 3000
     }
   ],
   "logging_steps": 50,
@@ -416,7 +494,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 4.477753664307475e+16,
+  "total_flos": 5.37491824204009e+16,
   "train_batch_size": 16,
   "trial_name": null,
   "trial_params": null