Training in progress, step 3500, checkpoint

last-checkpoint/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b39bfb04c37041840837102213b584e6ae1204eb3c1f47b35a8ac4f25c8411ea
+oid sha256:12ebe9f8ae824e04ec42c21281b6ee77ecf6ec560c2067b986f9a3a7439557b1
 size 577789320

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:55c130523c1814ac02576b7bcd8ddb8224d21012fa3dc0b1b8581ea711f9eefd
+oid sha256:0ff28dd97c1be0347d1839efcbc7062ce46057ca02707999acb5822aa7d44926
 size 1155772233

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b0bd26a66f18a63513b4644dff9bbce4420bc172da7cc36bbc9e2bbc11796d8c
+oid sha256:aac92226f0d21eb319c78dbae49877455a40d317910aac9583ab04de32de7d1c
 size 14244

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:390813e12fc409286b76f8ab397b98e6ba085504e2b094a343edebab876251b0
+oid sha256:1a4b47ac4e852aed8ea089749b4911c7dbd61b1aa03f00c4803df03fc43f36e5
 size 1064

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 0.3480445444583893,
-  "best_model_checkpoint": "mikhail_panzo/zlm_b128_le4_s8000/checkpoint-3000",
-  "epoch": 5.026178010471204,
+  "best_metric": 0.3431868255138397,
+  "best_model_checkpoint": "mikhail_panzo/zlm_b128_le4_s8000/checkpoint-3500",
+  "epoch": 5.863874345549738,
   "eval_steps": 500,
-  "global_step": 3000,
+  "global_step": 3500,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -475,6 +475,84 @@
       "eval_samples_per_second": 31.129,
       "eval_steps_per_second": 3.894,
       "step": 3000
+    },
+    {
+      "epoch": 5.109947643979058,
+      "grad_norm": 1.5680384635925293,
+      "learning_rate": 8.253333333333334e-05,
+      "loss": 0.3739,
+      "step": 3050
+    },
+    {
+      "epoch": 5.193717277486911,
+      "grad_norm": 1.5168578624725342,
+      "learning_rate": 8.17e-05,
+      "loss": 0.3756,
+      "step": 3100
+    },
+    {
+      "epoch": 5.277486910994765,
+      "grad_norm": 1.0925545692443848,
+      "learning_rate": 8.086666666666666e-05,
+      "loss": 0.3739,
+      "step": 3150
+    },
+    {
+      "epoch": 5.361256544502618,
+      "grad_norm": 0.8022084832191467,
+      "learning_rate": 8.003333333333333e-05,
+      "loss": 0.3736,
+      "step": 3200
+    },
+    {
+      "epoch": 5.445026178010472,
+      "grad_norm": 1.6964820623397827,
+      "learning_rate": 7.920000000000001e-05,
+      "loss": 0.379,
+      "step": 3250
+    },
+    {
+      "epoch": 5.528795811518324,
+      "grad_norm": 1.0717741250991821,
+      "learning_rate": 7.836666666666667e-05,
+      "loss": 0.3762,
+      "step": 3300
+    },
+    {
+      "epoch": 5.612565445026178,
+      "grad_norm": 1.2660809755325317,
+      "learning_rate": 7.753333333333334e-05,
+      "loss": 0.3708,
+      "step": 3350
+    },
+    {
+      "epoch": 5.696335078534031,
+      "grad_norm": 1.6065802574157715,
+      "learning_rate": 7.670000000000001e-05,
+      "loss": 0.3717,
+      "step": 3400
+    },
+    {
+      "epoch": 5.780104712041885,
+      "grad_norm": 1.2089744806289673,
+      "learning_rate": 7.586666666666668e-05,
+      "loss": 0.3715,
+      "step": 3450
+    },
+    {
+      "epoch": 5.863874345549738,
+      "grad_norm": 3.1524300575256348,
+      "learning_rate": 7.503333333333333e-05,
+      "loss": 0.3708,
+      "step": 3500
+    },
+    {
+      "epoch": 5.863874345549738,
+      "eval_loss": 0.3431868255138397,
+      "eval_runtime": 272.9384,
+      "eval_samples_per_second": 31.102,
+      "eval_steps_per_second": 3.891,
+      "step": 3500
     }
   ],
   "logging_steps": 50,
@@ -494,7 +572,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 5.37491824204009e+16,
+  "total_flos": 6.269975786347891e+16,
   "train_batch_size": 16,
   "trial_name": null,
   "trial_params": null