Training in progress, step 2400, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:01d3c25fe15711de727142e72fcf18b1562f70ba4803b2ccb81aa11f5be7f0b6
+oid sha256:22540877ddeb300d82f058269ceb097a438e294352fd79b189b18c52993ffece
 size 41977360

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1d9ffc1c556923224e5accb81e675f29bc7461c6ef6704644fe820ff7da2d25a
+oid sha256:08fa2f23217a52fb2ac6fb72c6821bb0c5bf1c7e6a71790e6b479176a783c832
 size 84100858

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4a00c887d1f42518acb45d72e988d69cde39eeaca6011972cd0e04b64ce69f72
+oid sha256:9a072144747ccaadd2037da23dfb15caa378253fefdb04cc465da01347494002
 size 14244

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d6dee65d23c30e8b90e557d338adbbb6ac407846ac98bf5125837e5e2dee92dc
+oid sha256:76f9460ae1eed76487f015151baa63c1c673e15bd3c50265cddbe79c4fc7df9c
 size 1064

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
   "best_metric": 0.7297702431678772,
   "best_model_checkpoint": "./results/phi2-mentalchat16k/checkpoint-2000",
-  "epoch": 3.2909498878085266,
+  "epoch": 3.590127150336574,
   "eval_steps": 100,
-  "global_step": 2200,
+  "global_step": 2400,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -498,6 +498,50 @@
       "eval_samples_per_second": 3.048,
       "eval_steps_per_second": 0.762,
       "step": 2200
+    },
+    {
+      "epoch": 3.3657442034405385,
+      "grad_norm": 0.3087999224662781,
+      "learning_rate": 1.2807482414650063e-05,
+      "loss": 0.6879,
+      "step": 2250
+    },
+    {
+      "epoch": 3.4405385190725504,
+      "grad_norm": 0.3264073431491852,
+      "learning_rate": 1.0001043028542834e-05,
+      "loss": 0.6867,
+      "step": 2300
+    },
+    {
+      "epoch": 3.4405385190725504,
+      "eval_loss": 0.7300211787223816,
+      "eval_runtime": 38.8364,
+      "eval_samples_per_second": 12.154,
+      "eval_steps_per_second": 3.038,
+      "step": 2300
+    },
+    {
+      "epoch": 3.5153328347045623,
+      "grad_norm": 0.35010454058647156,
+      "learning_rate": 7.5252847496027565e-06,
+      "loss": 0.6721,
+      "step": 2350
+    },
+    {
+      "epoch": 3.590127150336574,
+      "grad_norm": 0.31919124722480774,
+      "learning_rate": 5.389304199794209e-06,
+      "loss": 0.6817,
+      "step": 2400
+    },
+    {
+      "epoch": 3.590127150336574,
+      "eval_loss": 0.7298696637153625,
+      "eval_runtime": 38.8245,
+      "eval_samples_per_second": 12.157,
+      "eval_steps_per_second": 3.039,
+      "step": 2400
     }
   ],
   "logging_steps": 50,
@@ -512,7 +556,7 @@
         "early_stopping_threshold": 0.001
       },
       "attributes": {
-        "early_stopping_patience_counter": 2
+        "early_stopping_patience_counter": 4
       }
     },
     "TrainerControl": {
@@ -526,7 +570,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 1.341425978522112e+17,
+  "total_flos": 1.464313302280704e+17,
   "train_batch_size": 4,
   "trial_name": null,
   "trial_params": null