Training in progress, step 2000, checkpoint

last-checkpoint/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bb0a458b340d24f70fc214681d5c5119f47646e9547550f4a52c22623cc545e0
+oid sha256:543ba4376065b1459d1d1e92a4d4912f7658d805e431dda1e8d7076b636b8dec
 size 577789320

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3e145b91418be3b833420a93109497d7460e2e0c60c30711829d5ecce85f1097
+oid sha256:c5e89b6713085fb5d6ebc9f3653b88f811d1b5ec815e6e181c34a162bcbe7c59
 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:f32a07717606d664792d4ebfd434223710fe948a637ff5f34234da98aa96ac43
+oid sha256:b911007def71d4e0760e095fb38b9695588c94c6dfc0cb1ee70d933b50c5ece7
 size 1064

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 0.44668668508529663,
-  "best_model_checkpoint": "mikhail-panzo/zlm_b128_le5_s8000/checkpoint-1500",
-  "epoch": 2.513089005235602,
+  "best_metric": 0.4236186444759369,
+  "best_model_checkpoint": "mikhail-panzo/zlm_b128_le5_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": 33.638,
       "eval_steps_per_second": 4.208,
       "step": 1500
+    },
+    {
+      "epoch": 2.5968586387434556,
+      "grad_norm": 1.6376454830169678,
+      "learning_rate": 7.745e-06,
+      "loss": 0.503,
+      "step": 1550
+    },
+    {
+      "epoch": 2.680628272251309,
+      "grad_norm": 1.7104960680007935,
+      "learning_rate": 7.995e-06,
+      "loss": 0.4988,
+      "step": 1600
+    },
+    {
+      "epoch": 2.7643979057591626,
+      "grad_norm": 2.3372185230255127,
+      "learning_rate": 8.245000000000002e-06,
+      "loss": 0.4954,
+      "step": 1650
+    },
+    {
+      "epoch": 2.8481675392670156,
+      "grad_norm": 2.170474052429199,
+      "learning_rate": 8.495e-06,
+      "loss": 0.4971,
+      "step": 1700
+    },
+    {
+      "epoch": 2.931937172774869,
+      "grad_norm": 2.3041465282440186,
+      "learning_rate": 8.745000000000002e-06,
+      "loss": 0.4948,
+      "step": 1750
+    },
+    {
+      "epoch": 3.0157068062827226,
+      "grad_norm": 2.1798529624938965,
+      "learning_rate": 8.995000000000001e-06,
+      "loss": 0.4849,
+      "step": 1800
+    },
+    {
+      "epoch": 3.099476439790576,
+      "grad_norm": 1.8813133239746094,
+      "learning_rate": 9.245e-06,
+      "loss": 0.4839,
+      "step": 1850
+    },
+    {
+      "epoch": 3.183246073298429,
+      "grad_norm": 1.8749632835388184,
+      "learning_rate": 9.49e-06,
+      "loss": 0.4871,
+      "step": 1900
+    },
+    {
+      "epoch": 3.2670157068062826,
+      "grad_norm": 1.917149305343628,
+      "learning_rate": 9.74e-06,
+      "loss": 0.4724,
+      "step": 1950
+    },
+    {
+      "epoch": 3.350785340314136,
+      "grad_norm": 1.9401865005493164,
+      "learning_rate": 9.990000000000001e-06,
+      "loss": 0.4776,
+      "step": 2000
+    },
+    {
+      "epoch": 3.350785340314136,
+      "eval_loss": 0.4236186444759369,
+      "eval_runtime": 261.4044,
+      "eval_samples_per_second": 32.475,
+      "eval_steps_per_second": 4.063,
+      "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