Training in progress, step 4000, checkpoint

last-checkpoint/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:12ebe9f8ae824e04ec42c21281b6ee77ecf6ec560c2067b986f9a3a7439557b1
+oid sha256:09f46755ffeadc3d999dd23fd7022daf35bf2fbdd0db7cbfb92677a532f0a38f
 size 577789320

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0ff28dd97c1be0347d1839efcbc7062ce46057ca02707999acb5822aa7d44926
+oid sha256:4766a7bbe36c3cafb785ca7ab63a31dbb580385e53a5efb78b1082ff7c77f718
 size 1155772233

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:aac92226f0d21eb319c78dbae49877455a40d317910aac9583ab04de32de7d1c
+oid sha256:56cc8e2d77d54e9e58cdbed4bf8357f729e20efe110f0bdfcd24e4976aebb3b0
 size 14244

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1a4b47ac4e852aed8ea089749b4911c7dbd61b1aa03f00c4803df03fc43f36e5
+oid sha256:6f3323bc562f08c84a55f57ffe42cad4e899b9bed240a410d83f006b05ef914d
 size 1064

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 0.3431868255138397,
-  "best_model_checkpoint": "mikhail_panzo/zlm_b128_le4_s8000/checkpoint-3500",
-  "epoch": 5.863874345549738,
+  "best_metric": 0.33282962441444397,
+  "best_model_checkpoint": "mikhail_panzo/zlm_b128_le4_s8000/checkpoint-4000",
+  "epoch": 6.701570680628272,
   "eval_steps": 500,
-  "global_step": 3500,
+  "global_step": 4000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -553,6 +553,84 @@
       "eval_samples_per_second": 31.102,
       "eval_steps_per_second": 3.891,
       "step": 3500
+    },
+    {
+      "epoch": 5.947643979057592,
+      "grad_norm": 1.914294719696045,
+      "learning_rate": 7.42e-05,
+      "loss": 0.3673,
+      "step": 3550
+    },
+    {
+      "epoch": 6.031413612565445,
+      "grad_norm": 1.3186005353927612,
+      "learning_rate": 7.336666666666667e-05,
+      "loss": 0.3665,
+      "step": 3600
+    },
+    {
+      "epoch": 6.115183246073299,
+      "grad_norm": 2.019273519515991,
+      "learning_rate": 7.253333333333334e-05,
+      "loss": 0.3697,
+      "step": 3650
+    },
+    {
+      "epoch": 6.198952879581152,
+      "grad_norm": 1.0517597198486328,
+      "learning_rate": 7.17e-05,
+      "loss": 0.3674,
+      "step": 3700
+    },
+    {
+      "epoch": 6.282722513089006,
+      "grad_norm": 1.0202686786651611,
+      "learning_rate": 7.086666666666666e-05,
+      "loss": 0.3706,
+      "step": 3750
+    },
+    {
+      "epoch": 6.366492146596858,
+      "grad_norm": 1.4179818630218506,
+      "learning_rate": 7.003333333333335e-05,
+      "loss": 0.3681,
+      "step": 3800
+    },
+    {
+      "epoch": 6.450261780104712,
+      "grad_norm": 1.3820505142211914,
+      "learning_rate": 6.92e-05,
+      "loss": 0.3671,
+      "step": 3850
+    },
+    {
+      "epoch": 6.534031413612565,
+      "grad_norm": 1.3857202529907227,
+      "learning_rate": 6.836666666666667e-05,
+      "loss": 0.3641,
+      "step": 3900
+    },
+    {
+      "epoch": 6.617801047120419,
+      "grad_norm": 1.0996108055114746,
+      "learning_rate": 6.753333333333334e-05,
+      "loss": 0.3685,
+      "step": 3950
+    },
+    {
+      "epoch": 6.701570680628272,
+      "grad_norm": 0.9405946731567383,
+      "learning_rate": 6.670000000000001e-05,
+      "loss": 0.366,
+      "step": 4000
+    },
+    {
+      "epoch": 6.701570680628272,
+      "eval_loss": 0.33282962441444397,
+      "eval_runtime": 271.6112,
+      "eval_samples_per_second": 31.254,
+      "eval_steps_per_second": 3.91,
+      "step": 4000
     }
   ],
   "logging_steps": 50,
@@ -572,7 +650,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 6.269975786347891e+16,
+  "total_flos": 7.164514367145274e+16,
   "train_batch_size": 16,
   "trial_name": null,
   "trial_params": null