Training in progress, step 1000, checkpoint

last-checkpoint/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ad6aeef38ed358f8048c115091424d7e45cd8e98e79fdc324106be4910f58c95
+oid sha256:6b3bd15ad5f6e246b6e337225788c69c49a21df6eeb262d2b1b1897cf5cf6146
 size 4523108832

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b2d76978d088fcb730ac283219f0e32e79d7d521d55adaf8f7006b69f804ac5c
+oid sha256:9ecf8f056195b0098a8274b3035e24672255ff4c10ed92a0717c3ca84f2c7426
 size 2911851147

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:098b29492211804ab324a36f37466821d948280bb74fce4ba895c03f13ecd878
+oid sha256:a8e2011629d8bed3ef560fa11175cac55684c4e12a72634bb24abf767b6c7399
 size 14645

last-checkpoint/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f77569c2e850b04af982cc8c1389f1430851448915c593b69e5da36ce05b71d7
+oid sha256:14ae2a2128444abab378aa06c09a61a84665f758fcc19fc46f5789b0bc1b5665
 size 1383

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ff3b68c80bb6e1918e0fcbbfb26fa72db3f12e89d05c20abd2bd696373cfaadc
+oid sha256:92235ff997c3a5622fba1c344122ccffab641b57d7443401a19efb448a747750
 size 1465

last-checkpoint/trainer_state.json

@@ -1,10 +1,10 @@
 {
-  "best_global_step": 500,
-  "best_metric": 1.041561245918274,
-  "best_model_checkpoint": "./sft-out/checkpoint-500",
-  "epoch": 0.8840267418089397,
+  "best_global_step": 1000,
+  "best_metric": 0.988082230091095,
+  "best_model_checkpoint": "./sft-out/checkpoint-1000",
+  "epoch": 1.7673352118901597,
   "eval_steps": 500,
-  "global_step": 500,
+  "global_step": 1000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -86,6 +86,84 @@
       "eval_samples_per_second": 18.184,
       "eval_steps_per_second": 2.291,
       "step": 500
+    },
+    {
+      "epoch": 0.9724294159898337,
+      "grad_norm": 2.2783753871917725,
+      "learning_rate": 1.884018956444297e-05,
+      "loss": 1.068794708251953,
+      "step": 550
+    },
+    {
+      "epoch": 1.060113818443008,
+      "grad_norm": 2.0879323482513428,
+      "learning_rate": 1.85741517679431e-05,
+      "loss": 0.8203845977783203,
+      "step": 600
+    },
+    {
+      "epoch": 1.148516492623902,
+      "grad_norm": 2.290803909301758,
+      "learning_rate": 1.828296450700078e-05,
+      "loss": 0.7069022369384765,
+      "step": 650
+    },
+    {
+      "epoch": 1.2369191668047959,
+      "grad_norm": 2.6409525871276855,
+      "learning_rate": 1.7967481884023257e-05,
+      "loss": 0.707320556640625,
+      "step": 700
+    },
+    {
+      "epoch": 1.3253218409856897,
+      "grad_norm": 2.363614559173584,
+      "learning_rate": 1.7628629263900676e-05,
+      "loss": 0.7089004516601562,
+      "step": 750
+    },
+    {
+      "epoch": 1.4137245151665838,
+      "grad_norm": 2.3598053455352783,
+      "learning_rate": 1.7267400559751396e-05,
+      "loss": 0.709156265258789,
+      "step": 800
+    },
+    {
+      "epoch": 1.502127189347478,
+      "grad_norm": 2.0364632606506348,
+      "learning_rate": 1.6884855317603584e-05,
+      "loss": 0.6911470031738282,
+      "step": 850
+    },
+    {
+      "epoch": 1.5905298635283718,
+      "grad_norm": 2.0562117099761963,
+      "learning_rate": 1.648211560856415e-05,
+      "loss": 0.6989401245117187,
+      "step": 900
+    },
+    {
+      "epoch": 1.6789325377092656,
+      "grad_norm": 2.351238965988159,
+      "learning_rate": 1.6060362737590845e-05,
+      "loss": 0.7044075775146484,
+      "step": 950
+    },
+    {
+      "epoch": 1.7673352118901597,
+      "grad_norm": 2.1121199131011963,
+      "learning_rate": 1.5620833778521306e-05,
+      "loss": 0.7032404327392578,
+      "step": 1000
+    },
+    {
+      "epoch": 1.7673352118901597,
+      "eval_loss": 0.988082230091095,
+      "eval_runtime": 27.3518,
+      "eval_samples_per_second": 18.28,
+      "eval_steps_per_second": 2.303,
+      "step": 1000
     }
   ],
   "logging_steps": 50,
@@ -105,7 +183,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 1.9429307822186496e+16,
+  "total_flos": 3.876769443016704e+16,
   "train_batch_size": 1,
   "trial_name": null,
   "trial_params": null