Training in progress, step 2000, checkpoint

last-checkpoint/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9ff41608dfc78b9bdc4f8276043a06dc109f8d05bb59380643a3c730dc24b969
+oid sha256:17756aebf81d95bb9838128283e6da6710d776c7d9df7c685f664101b5b99452
 size 577789320

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cb8b164b44bfdcf123dbc3a7302020cababfdfb0a58f38d54366139f9f307231
+oid sha256:9b487bee88037fce1eca80b45882678fd5954a83a75d4060d95c1e16fccf145d
 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": 0.3777858018875122,
-  "best_model_checkpoint": "mikhail_panzo/zlm_b128_le4_s8000/checkpoint-1500",
-  "epoch": 2.513089005235602,
+  "best_metric": 0.3677983582019806,
+  "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": 31.058,
       "eval_steps_per_second": 3.885,
       "step": 1500
+    },
+    {
+      "epoch": 2.5968586387434556,
+      "grad_norm": 1.6483193635940552,
+      "learning_rate": 7.745e-05,
+      "loss": 0.414,
+      "step": 1550
+    },
+    {
+      "epoch": 2.680628272251309,
+      "grad_norm": 1.7688554525375366,
+      "learning_rate": 7.995e-05,
+      "loss": 0.4153,
+      "step": 1600
+    },
+    {
+      "epoch": 2.7643979057591626,
+      "grad_norm": 1.2314317226409912,
+      "learning_rate": 8.245e-05,
+      "loss": 0.4089,
+      "step": 1650
+    },
+    {
+      "epoch": 2.8481675392670156,
+      "grad_norm": 1.6623793840408325,
+      "learning_rate": 8.495e-05,
+      "loss": 0.4124,
+      "step": 1700
+    },
+    {
+      "epoch": 2.931937172774869,
+      "grad_norm": 3.812507390975952,
+      "learning_rate": 8.745000000000001e-05,
+      "loss": 0.4112,
+      "step": 1750
+    },
+    {
+      "epoch": 3.0157068062827226,
+      "grad_norm": 2.141019821166992,
+      "learning_rate": 8.995e-05,
+      "loss": 0.4081,
+      "step": 1800
+    },
+    {
+      "epoch": 3.099476439790576,
+      "grad_norm": 1.8928133249282837,
+      "learning_rate": 9.245e-05,
+      "loss": 0.4067,
+      "step": 1850
+    },
+    {
+      "epoch": 3.183246073298429,
+      "grad_norm": 2.322817087173462,
+      "learning_rate": 9.495e-05,
+      "loss": 0.4088,
+      "step": 1900
+    },
+    {
+      "epoch": 3.2670157068062826,
+      "grad_norm": 2.1984918117523193,
+      "learning_rate": 9.745000000000001e-05,
+      "loss": 0.3976,
+      "step": 1950
+    },
+    {
+      "epoch": 3.350785340314136,
+      "grad_norm": 2.0455121994018555,
+      "learning_rate": 9.995e-05,
+      "loss": 0.4022,
+      "step": 2000
+    },
+    {
+      "epoch": 3.350785340314136,
+      "eval_loss": 0.3677983582019806,
+      "eval_runtime": 274.4574,
+      "eval_samples_per_second": 30.93,
+      "eval_steps_per_second": 3.869,
+      "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