Training in progress, step 2000, checkpoint

last-checkpoint/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:266d344bebc5ab5ce942ef956f2aa9295d8e94a6a57cbc062cb783dd1e9232a9
+oid sha256:707a3dd059e6ef7b537429ab982116673e7c57856a2baabb5e0e93a37db1e05a
 size 577789320

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5cabaf1ffdfcf66c4597c58e5011be28caacdfbda28f043b6b27a7958894120e
+oid sha256:26d1e099ef331081180df0b687b1bbb9496a9e44f71a8084cf37a61c56a44eed
 size 1155772233

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fb8bd6e628183ec05339d8819c629745ae8132daa41aa7ccb3b448576240129f
+oid sha256:65e2559ac3f951b931d090db8ae4351b502c581ebb31bfce4ac34fe597db3515
 size 14244

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f32a07717606d664792d4ebfd434223710fe948a637ff5f34234da98aa96ac43
+oid sha256:8200c7911b70674e95f6ea3c4fd4f7c5a623c58e7efa3752996c3106e6a97146
 size 1064

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 0.43296581506729126,
-  "best_model_checkpoint": "mikhail_panzo/zlm-fil_b64_le5_s8000/checkpoint-1500",
-  "epoch": 65.21739130434783,
+  "best_metric": 0.42301931977272034,
+  "best_model_checkpoint": "mikhail_panzo/zlm-fil_b64_le5_s8000/checkpoint-2000",
+  "epoch": 86.95652173913044,
   "eval_steps": 500,
-  "global_step": 1500,
+  "global_step": 2000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -241,6 +241,84 @@
       "eval_samples_per_second": 12.521,
       "eval_steps_per_second": 1.633,
       "step": 1500
+    },
+    {
+      "epoch": 67.3913043478261,
+      "grad_norm": 1.6638849973678589,
+      "learning_rate": 7.745e-06,
+      "loss": 0.4674,
+      "step": 1550
+    },
+    {
+      "epoch": 69.56521739130434,
+      "grad_norm": 0.8030112981796265,
+      "learning_rate": 7.995e-06,
+      "loss": 0.4652,
+      "step": 1600
+    },
+    {
+      "epoch": 71.73913043478261,
+      "grad_norm": 1.2407337427139282,
+      "learning_rate": 8.245000000000002e-06,
+      "loss": 0.4638,
+      "step": 1650
+    },
+    {
+      "epoch": 73.91304347826087,
+      "grad_norm": 1.0331224203109741,
+      "learning_rate": 8.495e-06,
+      "loss": 0.4685,
+      "step": 1700
+    },
+    {
+      "epoch": 76.08695652173913,
+      "grad_norm": 2.4681615829467773,
+      "learning_rate": 8.745000000000002e-06,
+      "loss": 0.4627,
+      "step": 1750
+    },
+    {
+      "epoch": 78.26086956521739,
+      "grad_norm": 0.9344178438186646,
+      "learning_rate": 8.995000000000001e-06,
+      "loss": 0.4575,
+      "step": 1800
+    },
+    {
+      "epoch": 80.43478260869566,
+      "grad_norm": 1.0561341047286987,
+      "learning_rate": 9.245e-06,
+      "loss": 0.4621,
+      "step": 1850
+    },
+    {
+      "epoch": 82.6086956521739,
+      "grad_norm": 0.9816989302635193,
+      "learning_rate": 9.495000000000001e-06,
+      "loss": 0.4544,
+      "step": 1900
+    },
+    {
+      "epoch": 84.78260869565217,
+      "grad_norm": 1.4779256582260132,
+      "learning_rate": 9.745e-06,
+      "loss": 0.4579,
+      "step": 1950
+    },
+    {
+      "epoch": 86.95652173913044,
+      "grad_norm": 1.1634950637817383,
+      "learning_rate": 9.995000000000002e-06,
+      "loss": 0.4518,
+      "step": 2000
+    },
+    {
+      "epoch": 86.95652173913044,
+      "eval_loss": 0.42301931977272034,
+      "eval_runtime": 12.725,
+      "eval_samples_per_second": 12.652,
+      "eval_steps_per_second": 1.65,
+      "step": 2000
     }
   ],
   "logging_steps": 50,
@@ -260,7 +338,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 1.8173480279405616e+16,
+  "total_flos": 2.424791006072741e+16,
   "train_batch_size": 32,
   "trial_name": null,
   "trial_params": null