Training in progress, step 5500, checkpoint

last-checkpoint/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:35ee42eb255f39420e07c05702286cc97c0798609311d524893a0f1e9d7cc859
+oid sha256:10acedbc50fa3d8c8d9dbe4eb7fab990e97c4fcef731b7ef8aad5b6f0f36fc31
 size 577789320

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0e076514e9e3416400a9640a95979e16ac08d9c27980d9e454df36fbd79ae368
+oid sha256:ca17de2a3f967c4fbbb47ab1fd051b2406152d9e83655a9085db81c67f9ce41e
 size 1155772233

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d9139636e888f3e7631c5b3a77700c20686968ce93ccddda8fd771e1d5c792bc
+oid sha256:e5c35ebbbf890f698cdf40bc9eba9c8ea419b358821e4143706913f24529731a
 size 14244

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1b02450bf90416881debf85ba0d776babdc9d2efe4751382ccfff0b7db595003
+oid sha256:5b353b67d3c68734cb4cbcb55694f7eeeb40f54128ba041a6736d420b09008b6
 size 1064

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 0.40913259983062744,
-  "best_model_checkpoint": "mikhail_panzo/zlm-fil_b64_le5_s8000/checkpoint-4500",
-  "epoch": 217.3913043478261,
+  "best_metric": 0.4069381058216095,
+  "best_model_checkpoint": "mikhail_panzo/zlm-fil_b64_le5_s8000/checkpoint-5500",
+  "epoch": 239.1304347826087,
   "eval_steps": 500,
-  "global_step": 5000,
+  "global_step": 5500,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -787,6 +787,84 @@
       "eval_samples_per_second": 12.11,
       "eval_steps_per_second": 1.58,
       "step": 5000
+    },
+    {
+      "epoch": 219.56521739130434,
+      "grad_norm": 0.9140251874923706,
+      "learning_rate": 4.92e-06,
+      "loss": 0.4142,
+      "step": 5050
+    },
+    {
+      "epoch": 221.7391304347826,
+      "grad_norm": 0.7339205741882324,
+      "learning_rate": 4.836666666666667e-06,
+      "loss": 0.4126,
+      "step": 5100
+    },
+    {
+      "epoch": 223.91304347826087,
+      "grad_norm": 0.8861171007156372,
+      "learning_rate": 4.753333333333333e-06,
+      "loss": 0.4189,
+      "step": 5150
+    },
+    {
+      "epoch": 226.08695652173913,
+      "grad_norm": 0.7763547301292419,
+      "learning_rate": 4.670000000000001e-06,
+      "loss": 0.415,
+      "step": 5200
+    },
+    {
+      "epoch": 228.2608695652174,
+      "grad_norm": 0.913635790348053,
+      "learning_rate": 4.586666666666667e-06,
+      "loss": 0.413,
+      "step": 5250
+    },
+    {
+      "epoch": 230.43478260869566,
+      "grad_norm": 1.0177322626113892,
+      "learning_rate": 4.503333333333333e-06,
+      "loss": 0.4119,
+      "step": 5300
+    },
+    {
+      "epoch": 232.6086956521739,
+      "grad_norm": 1.0655298233032227,
+      "learning_rate": 4.42e-06,
+      "loss": 0.4122,
+      "step": 5350
+    },
+    {
+      "epoch": 234.7826086956522,
+      "grad_norm": 0.84236079454422,
+      "learning_rate": 4.336666666666667e-06,
+      "loss": 0.4104,
+      "step": 5400
+    },
+    {
+      "epoch": 236.95652173913044,
+      "grad_norm": 1.0191460847854614,
+      "learning_rate": 4.253333333333334e-06,
+      "loss": 0.4128,
+      "step": 5450
+    },
+    {
+      "epoch": 239.1304347826087,
+      "grad_norm": 0.7806400060653687,
+      "learning_rate": 4.17e-06,
+      "loss": 0.414,
+      "step": 5500
+    },
+    {
+      "epoch": 239.1304347826087,
+      "eval_loss": 0.4069381058216095,
+      "eval_runtime": 13.5672,
+      "eval_samples_per_second": 11.867,
+      "eval_steps_per_second": 1.548,
+      "step": 5500
     }
   ],
   "logging_steps": 50,
@@ -806,7 +884,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 6.05935368348156e+16,
+  "total_flos": 6.665726342678198e+16,
   "train_batch_size": 32,
   "trial_name": null,
   "trial_params": null