Training in progress, step 2400, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1ae6025af8600a07d10771669ed9775cfcce6fd2665c97ec98fe27b309d3a6a8
+oid sha256:efee5f15a6ab6ae2aac41a5b72e2348f297a8d9be317eefa6b81d4bf46f9b2d2
 size 50503544

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:daf7ee8a11f1f1dba3d9bd654e63f53102066bd7e502c39d4355f7ba2c78d561
+oid sha256:1a38b06fdba3de21eaaf685de4cad02d425e8664279937b12675acddcb15d2b6
 size 101184122

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3c80148c1a0ea492aeade85b5c4e3f4cac348e52120402746218f740878e1cfa
+oid sha256:05b18c2c337e77a16f945d65a90b4cc2762226e8c3d903c5ec5bf0d98bc25dba
 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": 0.7816067934036255,
   "best_model_checkpoint": "./results/tinyllama-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": 25.441,
       "eval_steps_per_second": 6.36,
       "step": 2200
+    },
+    {
+      "epoch": 3.3657442034405385,
+      "grad_norm": 0.6838712096214294,
+      "learning_rate": 1.2807482414650063e-05,
+      "loss": 0.529,
+      "step": 2250
+    },
+    {
+      "epoch": 3.4405385190725504,
+      "grad_norm": 0.7482870817184448,
+      "learning_rate": 1.0001043028542834e-05,
+      "loss": 0.5376,
+      "step": 2300
+    },
+    {
+      "epoch": 3.4405385190725504,
+      "eval_loss": 0.8013114929199219,
+      "eval_runtime": 18.5497,
+      "eval_samples_per_second": 25.445,
+      "eval_steps_per_second": 6.361,
+      "step": 2300
+    },
+    {
+      "epoch": 3.5153328347045623,
+      "grad_norm": 0.7258043885231018,
+      "learning_rate": 7.5252847496027565e-06,
+      "loss": 0.531,
+      "step": 2350
+    },
+    {
+      "epoch": 3.590127150336574,
+      "grad_norm": 0.6967834830284119,
+      "learning_rate": 5.389304199794209e-06,
+      "loss": 0.5365,
+      "step": 2400
+    },
+    {
+      "epoch": 3.590127150336574,
+      "eval_loss": 0.8018017411231995,
+      "eval_runtime": 18.5459,
+      "eval_samples_per_second": 25.45,
+      "eval_steps_per_second": 6.363,
+      "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": 5.564597695561728e+16,
+  "total_flos": 6.072984234550886e+16,
   "train_batch_size": 4,
   "trial_name": null,
   "trial_params": null