Training in progress, step 2500, checkpoint

last-checkpoint/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9fa58f85c1881fbdc1256be3df4827baa99f4d296a0d35ff985043a310ce6161
+oid sha256:369389b89182d14c0a29b5ed5a3b4212aa17dc2b6388dac028871806e82fc85b
 size 4523108832

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b20c0e007837bb2be3064494728543e31c81ffcdd62114afa5da005a51dd81cb
+oid sha256:681fe753a8832a47e6100d465b8935d28ed9dbae18200c385ba40d50b29c1454
 size 2911851147

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fdc37bbd2e979f041dfbbb004a5c74bab6cdda159cb18116df728588515a9ef6
+oid sha256:17cd930da9783ca70bad4b9cdeee6a06c0acea8f34645a333c93341f487f66a3
 size 14645

last-checkpoint/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f4aa03f6e0cd07cf67ce1fbe3101d545f5771ef9148b9debf02b11cf6948da5c
+oid sha256:861ce13e6ca091acee9a68ebfc5ca38479baf4b537c37b3949f071f77b81e9f0
 size 1383

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:14c891429705c32a3827fce3cd46c59925610cf516324635d98d8a689f74ce32
+oid sha256:65542a14b261e86856294f82b4572867fb3fbe696f9c1e5b6cf0bc6338925485
 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": 3.5339521520525996,
+  "epoch": 4.41726062213382,
   "eval_steps": 500,
-  "global_step": 2000,
+  "global_step": 2500,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -320,6 +320,84 @@
       "eval_samples_per_second": 18.24,
       "eval_steps_per_second": 2.298,
       "step": 2000
+    },
+    {
+      "epoch": 3.6223548262334937,
+      "grad_norm": 1.6168675422668457,
+      "learning_rate": 4.853746852293582e-06,
+      "loss": 0.13220906257629395,
+      "step": 2050
+    },
+    {
+      "epoch": 3.7107575004143873,
+      "grad_norm": 1.501587152481079,
+      "learning_rate": 4.397099057116449e-06,
+      "loss": 0.13320664405822755,
+      "step": 2100
+    },
+    {
+      "epoch": 3.7991601745952814,
+      "grad_norm": 1.7080086469650269,
+      "learning_rate": 3.956885535772039e-06,
+      "loss": 0.13129110336303712,
+      "step": 2150
+    },
+    {
+      "epoch": 3.8875628487761755,
+      "grad_norm": 1.4526396989822388,
+      "learning_rate": 3.534397510366242e-06,
+      "loss": 0.1242832088470459,
+      "step": 2200
+    },
+    {
+      "epoch": 3.975965522957069,
+      "grad_norm": 1.5429104566574097,
+      "learning_rate": 3.1308742110713165e-06,
+      "loss": 0.1259861946105957,
+      "step": 2250
+    },
+    {
+      "epoch": 4.063649925410243,
+      "grad_norm": 0.9340503811836243,
+      "learning_rate": 2.7474992412502555e-06,
+      "loss": 0.07559285640716552,
+      "step": 2300
+    },
+    {
+      "epoch": 4.152052599591138,
+      "grad_norm": 1.0590145587921143,
+      "learning_rate": 2.385397105744187e-06,
+      "loss": 0.05158629417419434,
+      "step": 2350
+    },
+    {
+      "epoch": 4.240455273772032,
+      "grad_norm": 1.2688530683517456,
+      "learning_rate": 2.04562991250593e-06,
+      "loss": 0.0525316333770752,
+      "step": 2400
+    },
+    {
+      "epoch": 4.328857947952925,
+      "grad_norm": 1.6087177991867065,
+      "learning_rate": 1.7291942572543806e-06,
+      "loss": 0.049799141883850095,
+      "step": 2450
+    },
+    {
+      "epoch": 4.41726062213382,
+      "grad_norm": 0.8664147853851318,
+      "learning_rate": 1.437018300287546e-06,
+      "loss": 0.05318857669830322,
+      "step": 2500
+    },
+    {
+      "epoch": 4.41726062213382,
+      "eval_loss": 1.2274935245513916,
+      "eval_runtime": 27.4265,
+      "eval_samples_per_second": 18.231,
+      "eval_steps_per_second": 2.297,
+      "step": 2500
     }
   ],
   "logging_steps": 50,
@@ -339,7 +417,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 7.762188828042854e+16,
+  "total_flos": 9.703744968828518e+16,
   "train_batch_size": 1,
   "trial_name": null,
   "trial_params": null