Model save

README.md

@@ -5,7 +5,6 @@ license: llama3
 tags:
 - trl
 - sft
-- alignment-handbook
 - generated_from_trainer
 model-index:
 - name: zephyr-7b-sft-qlora
@@ -41,8 +40,10 @@ The following hyperparameters were used during training:
 - eval_batch_size: 8
 - seed: 42
 - distributed_type: multi-GPU
+- num_devices: 4
 - gradient_accumulation_steps: 2
-- total_train_batch_size: 4
+- total_train_batch_size: 16
+- total_eval_batch_size: 32
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: cosine
 - lr_scheduler_warmup_ratio: 0.1

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2b9c2d65b260afb83795bef344ff2ec9cbb4279eadef0d113911fcc763f083da
+oid sha256:84bada5e93ed3642b542ab323222cd4b9749fed3285d4066aca495a25d8bcd48
 size 42002584

all_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 0.49,
-    "train_loss": 0.0,
-    "train_runtime": 0.0369,
+    "epoch": 1.0,
+    "train_loss": 0.6069550340240066,
+    "train_runtime": 2548.2523,
     "train_samples": 5919,
-    "train_samples_per_second": 160249.969,
-    "train_steps_per_second": 40069.261
+    "train_samples_per_second": 2.323,
+    "train_steps_per_second": 0.145
 }

runs/Aug02_02-10-21_21822c03bae0/events.out.tfevents.1722593430.21822c03bae0.34796.0

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0e4703c96ae940963b1e11d7ed7a00a7a15f289ee8afe1fc05c239d1737b4616
-size 20154
+oid sha256:09e143fcda35491ef2ef2885364173b30c6f5a1d164edf9de5e99f67fa31c720
+size 21352

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 0.49,
-    "train_loss": 0.0,
-    "train_runtime": 0.0369,
+    "epoch": 1.0,
+    "train_loss": 0.6069550340240066,
+    "train_runtime": 2548.2523,
     "train_samples": 5919,
-    "train_samples_per_second": 160249.969,
-    "train_steps_per_second": 40069.261
+    "train_samples_per_second": 2.323,
+    "train_steps_per_second": 0.145
 }