Training in progress, step 8000, checkpoint

last-checkpoint/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d17f4ae20f9f992a27e6a1fa6cb56310739b4d9e112f20feed6499f8e80bdb75
+oid sha256:55dbef2e4e7913ed689dee8375c55fdf478657367c0213dc4f662c92e2532c07
 size 577789320

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b092bb5f421e4613aec93ddde296b498489d2420e7c98850d80363bdeebccfc8
+oid sha256:4310d344d08e1b4d8a04135265c7e1212b3c9cacfe2d5d77dcb8fb7da2129ffd
 size 1155772233

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d880acef54d4eb61cf44bc2962e1f38bf44896cefc09a0303d9df0b77a1ae5c2
+oid sha256:012ecd1edf495661a426a7c3c4b090d5e2901f554832bf1d246a4e516cd8c109
 size 14244

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b1e6ef080b2d27380abb32e34a8b58f3a1c18cafb574b45dd697ad5ba8ccc473
+oid sha256:c517c807e8c60e96ea2a00ec87951911c3e0f81982c0a86873d457f034e52fcd
 size 1064

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 0.31849026679992676,
-  "best_model_checkpoint": "mikhail_panzo/zlm_b128_le4_s8000/checkpoint-7000",
-  "epoch": 12.565445026178011,
+  "best_metric": 0.31487980484962463,
+  "best_model_checkpoint": "mikhail_panzo/zlm_b128_le4_s8000/checkpoint-8000",
+  "epoch": 13.403141361256544,
   "eval_steps": 500,
-  "global_step": 7500,
+  "global_step": 8000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -1177,6 +1177,84 @@
       "eval_samples_per_second": 30.034,
       "eval_steps_per_second": 3.757,
       "step": 7500
+    },
+    {
+      "epoch": 12.649214659685864,
+      "grad_norm": 1.4210392236709595,
+      "learning_rate": 4.453e-05,
+      "loss": 0.3436,
+      "step": 7550
+    },
+    {
+      "epoch": 12.732984293193716,
+      "grad_norm": 1.275467038154602,
+      "learning_rate": 4.4030000000000004e-05,
+      "loss": 0.3453,
+      "step": 7600
+    },
+    {
+      "epoch": 12.81675392670157,
+      "grad_norm": 1.1207870244979858,
+      "learning_rate": 4.3530000000000005e-05,
+      "loss": 0.3438,
+      "step": 7650
+    },
+    {
+      "epoch": 12.900523560209423,
+      "grad_norm": 1.8535631895065308,
+      "learning_rate": 4.3030000000000006e-05,
+      "loss": 0.3442,
+      "step": 7700
+    },
+    {
+      "epoch": 12.984293193717278,
+      "grad_norm": 1.0426372289657593,
+      "learning_rate": 4.253e-05,
+      "loss": 0.3494,
+      "step": 7750
+    },
+    {
+      "epoch": 13.06806282722513,
+      "grad_norm": 1.3337020874023438,
+      "learning_rate": 4.203e-05,
+      "loss": 0.3413,
+      "step": 7800
+    },
+    {
+      "epoch": 13.151832460732985,
+      "grad_norm": 1.017905592918396,
+      "learning_rate": 4.1530000000000004e-05,
+      "loss": 0.3417,
+      "step": 7850
+    },
+    {
+      "epoch": 13.235602094240837,
+      "grad_norm": 1.166343331336975,
+      "learning_rate": 4.103e-05,
+      "loss": 0.3443,
+      "step": 7900
+    },
+    {
+      "epoch": 13.319371727748692,
+      "grad_norm": 1.4170418977737427,
+      "learning_rate": 4.053e-05,
+      "loss": 0.3433,
+      "step": 7950
+    },
+    {
+      "epoch": 13.403141361256544,
+      "grad_norm": 1.125741720199585,
+      "learning_rate": 4.003e-05,
+      "loss": 0.3422,
+      "step": 8000
+    },
+    {
+      "epoch": 13.403141361256544,
+      "eval_loss": 0.31487980484962463,
+      "eval_runtime": 278.3852,
+      "eval_samples_per_second": 30.494,
+      "eval_steps_per_second": 3.815,
+      "step": 8000
     }
   ],
   "logging_steps": 50,
@@ -1196,7 +1274,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 1.3437364892618304e+17,
+  "total_flos": 1.4332007404692806e+17,
   "train_batch_size": 16,
   "trial_name": null,
   "trial_params": null