Training in progress, epoch 1, checkpoint

checkpoint-212/README.md

@@ -17,5 +17,6 @@ The following `bitsandbytes` quantization config was used during training:
 - bnb_4bit_compute_dtype: float32
 ### Framework versions
 
+- PEFT 0.5.0
 
 - PEFT 0.5.0

checkpoint-212/adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:de8ce2bccb319d2bdfcd877592856a532a0375a6fdb46c6f0ffaa639260d8d42
-size 115364125
+oid sha256:5b378a4cd3d139e515bd4e2807512ea45d113d342988bfa581f3231540438a41
+size 117304221

checkpoint-212/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5fbb85a32e61c7d4d1e1021c86fe0c961f148308dfdf0e396e22686d9856e5b3
-size 230776198
+oid sha256:9a4765b3ef667428fdfefbb618186cc85331baa01cd9b550d6b3739d848ff945
+size 234652550

checkpoint-212/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:605c05828269613cb5866707586c2c342dd863c38ccde072ff12613b969d5903
+oid sha256:7c14f53adc12a7d09e088aa739d3d678cdfee4d7e810d4db1cbc2d9799383405
 size 627

checkpoint-212/trainer_state.json

@@ -10,16 +10,16 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "learning_rate": 0.000199,
-      "loss": 3.2614,
+      "learning_rate": 9.950235849056605e-06,
+      "loss": 6.539,
       "step": 212
     },
     {
       "epoch": 1.0,
-      "eval_loss": NaN,
-      "eval_runtime": 4.1946,
-      "eval_samples_per_second": 161.161,
-      "eval_steps_per_second": 20.264,
+      "eval_loss": 6.494007587432861,
+      "eval_runtime": 2.5292,
+      "eval_samples_per_second": 267.281,
+      "eval_steps_per_second": 33.608,
       "step": 212
     }
   ],

checkpoint-212/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:771d084b8daaadfbfc448be3f4420adca931f8c6583bdc3f881ade3e2af8ca6d
+oid sha256:932083cda874171956e29a9559decb3e85350d37f61015836b2c832b21575f2e
 size 4155