Training in progress, step 2000, checkpoint

last-checkpoint/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0b3878b2837c54460760b7689511b1b119ade195a40bcc1bd58d5a2fa1c333cc
+oid sha256:7308e6252b959616b2fe7c9a1e30cbe6a34ffb2d029f50db793552fbd2bdbfbc
 size 577789320

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:15856761cc33aa3ef4082afef8bcd92b764bb43930646ef8255717cfd74acefa
+oid sha256:0e080c2a47685ca953b3b9d5095ee4227f7d627920903f3549b23b13a6c86e24
 size 1155772233

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:dc65651f7b378ef8fe4da6f30ced75150f8b4133aff002554882ce4b7f48b94e
+oid sha256:2f01983801a45234478c13db780514a852cdeaff2aa79f279442e47dc68cb11d
 size 14244

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1a94e74103068a5fb8b31be8524087df4f57bd43f07d1db2b9d36084882fdbb5
+oid sha256:15b560a9dc43b2ed5d8c7b0910cf19c12068f6e2db4cd26fd270e940d4d1787b
 size 1064

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 0.3853827118873596,
-  "best_model_checkpoint": "mikhail_panzo/zlm_b128_le4_s8000/checkpoint-1500",
-  "epoch": 2.513089005235602,
+  "best_metric": 0.36490994691848755,
+  "best_model_checkpoint": "mikhail_panzo/zlm_b128_le4_s8000/checkpoint-2000",
+  "epoch": 3.350785340314136,
   "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": 31.527,
       "eval_steps_per_second": 3.944,
       "step": 1500
+    },
+    {
+      "epoch": 2.5968586387434556,
+      "grad_norm": 1.4221985340118408,
+      "learning_rate": 7.745e-05,
+      "loss": 0.4145,
+      "step": 1550
+    },
+    {
+      "epoch": 2.680628272251309,
+      "grad_norm": 3.651655435562134,
+      "learning_rate": 7.995e-05,
+      "loss": 0.4129,
+      "step": 1600
+    },
+    {
+      "epoch": 2.7643979057591626,
+      "grad_norm": 1.2700576782226562,
+      "learning_rate": 8.245e-05,
+      "loss": 0.4106,
+      "step": 1650
+    },
+    {
+      "epoch": 2.8481675392670156,
+      "grad_norm": 1.8690059185028076,
+      "learning_rate": 8.495e-05,
+      "loss": 0.4134,
+      "step": 1700
+    },
+    {
+      "epoch": 2.931937172774869,
+      "grad_norm": 2.483203649520874,
+      "learning_rate": 8.745000000000001e-05,
+      "loss": 0.413,
+      "step": 1750
+    },
+    {
+      "epoch": 3.0157068062827226,
+      "grad_norm": 1.738501787185669,
+      "learning_rate": 8.995e-05,
+      "loss": 0.4063,
+      "step": 1800
+    },
+    {
+      "epoch": 3.099476439790576,
+      "grad_norm": 1.837342619895935,
+      "learning_rate": 9.245e-05,
+      "loss": 0.4049,
+      "step": 1850
+    },
+    {
+      "epoch": 3.183246073298429,
+      "grad_norm": 1.284974455833435,
+      "learning_rate": 9.495e-05,
+      "loss": 0.4065,
+      "step": 1900
+    },
+    {
+      "epoch": 3.2670157068062826,
+      "grad_norm": 2.637281656265259,
+      "learning_rate": 9.745000000000001e-05,
+      "loss": 0.3986,
+      "step": 1950
+    },
+    {
+      "epoch": 3.350785340314136,
+      "grad_norm": 1.4775941371917725,
+      "learning_rate": 9.995e-05,
+      "loss": 0.4054,
+      "step": 2000
+    },
+    {
+      "epoch": 3.350785340314136,
+      "eval_loss": 0.36490994691848755,
+      "eval_runtime": 265.9165,
+      "eval_samples_per_second": 31.924,
+      "eval_steps_per_second": 3.994,
+      "step": 2000
     }
   ],
   "logging_steps": 50,
@@ -260,7 +338,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 2.686672014814656e+16,
+  "total_flos": 3.5804068604023104e+16,
   "train_batch_size": 16,
   "trial_name": null,
   "trial_params": null