Training in progress, step 2000, checkpoint

last-checkpoint/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f48d4e3b236d8f6efbf06705c5921fc52712f6cbe83333940faee84b86dc9b95
+oid sha256:9fa58f85c1881fbdc1256be3df4827baa99f4d296a0d35ff985043a310ce6161
 size 4523108832

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:890a01f7b0ed19d87aa41af9bd25032b9358906ea2c970f07da2b1cfdf970382
+oid sha256:b20c0e007837bb2be3064494728543e31c81ffcdd62114afa5da005a51dd81cb
 size 2911851147

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:01f9a0f7843a37be87edd23f4e88aa93b38b95cc2c07503eeb1cf2e4632453a2
+oid sha256:fdc37bbd2e979f041dfbbb004a5c74bab6cdda159cb18116df728588515a9ef6
 size 14645

last-checkpoint/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ca372268f4fa9335030c0cb7aedb6cdba75f457da50e7a4034abb1a2d0843689
+oid sha256:f4aa03f6e0cd07cf67ce1fbe3101d545f5771ef9148b9debf02b11cf6948da5c
 size 1383

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d87adb2ba2eff80e48f285a6c3b50b12e8fa43fde329e8b0d491370e73a397d0
+oid sha256:14c891429705c32a3827fce3cd46c59925610cf516324635d98d8a689f74ce32
 size 1465

last-checkpoint/trainer_state.json

@@ -2,9 +2,9 @@
   "best_global_step": 1000,
   "best_metric": 0.988082230091095,
   "best_model_checkpoint": "./sft-out/checkpoint-1000",
-  "epoch": 2.6506436819713795,
+  "epoch": 3.5339521520525996,
   "eval_steps": 500,
-  "global_step": 1500,
+  "global_step": 2000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -242,6 +242,84 @@
       "eval_samples_per_second": 18.279,
       "eval_steps_per_second": 2.303,
       "step": 1500
+    },
+    {
+      "epoch": 2.7390463561522735,
+      "grad_norm": 1.9865392446517944,
+      "learning_rate": 1.001083307599696e-05,
+      "loss": 0.34075843811035156,
+      "step": 1550
+    },
+    {
+      "epoch": 2.8274490303331676,
+      "grad_norm": 2.692328691482544,
+      "learning_rate": 9.469428420178699e-06,
+      "loss": 0.3529329299926758,
+      "step": 1600
+    },
+    {
+      "epoch": 2.9158517045140613,
+      "grad_norm": 2.0326497554779053,
+      "learning_rate": 8.92958002222035e-06,
+      "loss": 0.35016448974609377,
+      "step": 1650
+    },
+    {
+      "epoch": 3.003536106967236,
+      "grad_norm": 1.2595773935317993,
+      "learning_rate": 8.392871350487781e-06,
+      "loss": 0.3463246154785156,
+      "step": 1700
+    },
+    {
+      "epoch": 3.0919387811481296,
+      "grad_norm": 1.5270416736602783,
+      "learning_rate": 7.860876663988886e-06,
+      "loss": 0.14206992149353026,
+      "step": 1750
+    },
+    {
+      "epoch": 3.1803414553290237,
+      "grad_norm": 1.808855652809143,
+      "learning_rate": 7.335156394800651e-06,
+      "loss": 0.14106608390808106,
+      "step": 1800
+    },
+    {
+      "epoch": 3.2687441295099178,
+      "grad_norm": 1.834566593170166,
+      "learning_rate": 6.817252571053034e-06,
+      "loss": 0.138691987991333,
+      "step": 1850
+    },
+    {
+      "epoch": 3.357146803690812,
+      "grad_norm": 1.779943823814392,
+      "learning_rate": 6.308684293894747e-06,
+      "loss": 0.13664389610290528,
+      "step": 1900
+    },
+    {
+      "epoch": 3.4455494778717055,
+      "grad_norm": 1.5618979930877686,
+      "learning_rate": 5.810943281707862e-06,
+      "loss": 0.13606584548950196,
+      "step": 1950
+    },
+    {
+      "epoch": 3.5339521520525996,
+      "grad_norm": 1.8645941019058228,
+      "learning_rate": 5.325489494640788e-06,
+      "loss": 0.13294898033142089,
+      "step": 2000
+    },
+    {
+      "epoch": 3.5339521520525996,
+      "eval_loss": 1.1485551595687866,
+      "eval_runtime": 27.413,
+      "eval_samples_per_second": 18.24,
+      "eval_steps_per_second": 2.298,
+      "step": 2000
     }
   ],
   "logging_steps": 50,
@@ -261,7 +339,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 5.822502544962355e+16,
+  "total_flos": 7.762188828042854e+16,
   "train_batch_size": 1,
   "trial_name": null,
   "trial_params": null