Training in progress, step 7500, checkpoint

last-checkpoint/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cce8730ec7977cf6faded9c8a64710dd40c64e82312111fdae82ba0e37b6fe02
+oid sha256:f03919594e15aa04dbb8008cf60a2a2d5b89be3490f892590197689b8410a661
 size 577789320

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7d0bacbf128b4eb958fabc358356f531f65f6bd424a2a9be5add80fa30ab7cff
+oid sha256:77e4c693ccaf9249bd98af9f184a00d5eb2f36b77cb0e433fbc26661d80daa72
 size 1155772233

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a59b056016e351429f719aaf02cc6fa4544a2d92d2a3d69beeeb56674b12a1f2
+oid sha256:d880acef54d4eb61cf44bc2962e1f38bf44896cefc09a0303d9df0b77a1ae5c2
 size 14244

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a48d9034d2ce2771f0a840f8364e39645d6f213d3858f7f5342b741dc49975d2
+oid sha256:b5cdfe5ed4f14bdadfee62402701e9c3c91a7e1b8246c7c7f0be536b67574fb3
 size 1064

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 0.31611359119415283,
-  "best_model_checkpoint": "mikhail_panzo/zlm_b128_le4_s8000/checkpoint-7000",
-  "epoch": 11.727748691099476,
+  "best_metric": 0.3158954679965973,
+  "best_model_checkpoint": "mikhail_panzo/zlm_b128_le4_s8000/checkpoint-7500",
+  "epoch": 12.565445026178011,
   "eval_steps": 500,
-  "global_step": 7000,
+  "global_step": 7500,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -1099,6 +1099,84 @@
       "eval_samples_per_second": 31.262,
       "eval_steps_per_second": 3.911,
       "step": 7000
+    },
+    {
+      "epoch": 11.81151832460733,
+      "grad_norm": 0.9007428288459778,
+      "learning_rate": 1.586666666666667e-05,
+      "loss": 0.3461,
+      "step": 7050
+    },
+    {
+      "epoch": 11.895287958115183,
+      "grad_norm": 1.1172393560409546,
+      "learning_rate": 1.5033333333333336e-05,
+      "loss": 0.3489,
+      "step": 7100
+    },
+    {
+      "epoch": 11.979057591623036,
+      "grad_norm": 1.0454591512680054,
+      "learning_rate": 1.42e-05,
+      "loss": 0.3429,
+      "step": 7150
+    },
+    {
+      "epoch": 12.06282722513089,
+      "grad_norm": 1.1494097709655762,
+      "learning_rate": 1.3383333333333335e-05,
+      "loss": 0.3523,
+      "step": 7200
+    },
+    {
+      "epoch": 12.146596858638743,
+      "grad_norm": 0.8459360003471375,
+      "learning_rate": 1.255e-05,
+      "loss": 0.3469,
+      "step": 7250
+    },
+    {
+      "epoch": 12.230366492146597,
+      "grad_norm": 1.034494161605835,
+      "learning_rate": 1.1716666666666667e-05,
+      "loss": 0.3451,
+      "step": 7300
+    },
+    {
+      "epoch": 12.31413612565445,
+      "grad_norm": 1.1340311765670776,
+      "learning_rate": 1.0883333333333335e-05,
+      "loss": 0.3441,
+      "step": 7350
+    },
+    {
+      "epoch": 12.397905759162304,
+      "grad_norm": 1.4063786268234253,
+      "learning_rate": 1.005e-05,
+      "loss": 0.3472,
+      "step": 7400
+    },
+    {
+      "epoch": 12.481675392670157,
+      "grad_norm": 1.271904468536377,
+      "learning_rate": 9.216666666666666e-06,
+      "loss": 0.3458,
+      "step": 7450
+    },
+    {
+      "epoch": 12.565445026178011,
+      "grad_norm": 1.2097103595733643,
+      "learning_rate": 8.383333333333333e-06,
+      "loss": 0.3443,
+      "step": 7500
+    },
+    {
+      "epoch": 12.565445026178011,
+      "eval_loss": 0.3158954679965973,
+      "eval_runtime": 271.3497,
+      "eval_samples_per_second": 31.284,
+      "eval_steps_per_second": 3.914,
+      "step": 7500
     }
   ],
   "logging_steps": 50,
@@ -1118,7 +1196,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 1.2541009593096864e+17,
+  "total_flos": 1.3437364892618304e+17,
   "train_batch_size": 16,
   "trial_name": null,
   "trial_params": null