Model save

README.md

@@ -1,11 +1,10 @@
 ---
 base_model: meta-llama/Meta-Llama-3-8B
 datasets:
-- llama-duo/synth_classification_dataset_dedup
+- generator
 library_name: peft
 license: llama3
 tags:
-- alignment-handbook
 - trl
 - sft
 - generated_from_trainer
@@ -19,9 +18,9 @@ should probably proofread and complete it, then remove this comment. -->
 
 # llama3-8b-classification-gpt4o-100k
 
-This model is a fine-tuned version of [meta-llama/Meta-Llama-3-8B](https://huggingface.co/meta-llama/Meta-Llama-3-8B) on the llama-duo/synth_classification_dataset_dedup dataset.
+This model is a fine-tuned version of [meta-llama/Meta-Llama-3-8B](https://huggingface.co/meta-llama/Meta-Llama-3-8B) on the generator dataset.
 It achieves the following results on the evaluation set:
-- Loss: 1.8087
+- Loss: 2.0032
 
 ## Model description
 
@@ -42,27 +41,32 @@ More information needed
 The following hyperparameters were used during training:
 - learning_rate: 0.0002
 - train_batch_size: 4
-- eval_batch_size: 2
+- eval_batch_size: 4
 - seed: 42
 - distributed_type: multi-GPU
 - num_devices: 4
 - gradient_accumulation_steps: 2
 - total_train_batch_size: 32
-- total_eval_batch_size: 8
+- total_eval_batch_size: 16
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: cosine
 - lr_scheduler_warmup_ratio: 0.1
-- num_epochs: 5
+- num_epochs: 10
 
 ### Training results
 
-| Training Loss | Epoch  | Step | Validation Loss |
-|:-------------:|:------:|:----:|:---------------:|
-| 1.4429        | 0.9978 | 225  | 1.7484          |
-| 1.3594        | 2.0    | 451  | 1.7719          |
-| 1.3154        | 2.9978 | 676  | 1.7679          |
-| 1.268         | 4.0    | 902  | 1.7960          |
-| 1.2695        | 4.9889 | 1125 | 1.8087          |
+| Training Loss | Epoch | Step | Validation Loss |
+|:-------------:|:-----:|:----:|:---------------:|
+| 0.9183        | 1.0   | 237  | 1.6517          |
+| 0.8583        | 2.0   | 474  | 1.6295          |
+| 0.8179        | 3.0   | 711  | 1.6559          |
+| 0.7533        | 4.0   | 948  | 1.6894          |
+| 0.716         | 5.0   | 1185 | 1.7251          |
+| 0.6876        | 6.0   | 1422 | 1.7830          |
+| 0.6344        | 7.0   | 1659 | 1.8557          |
+| 0.591         | 8.0   | 1896 | 1.9240          |
+| 0.5677        | 9.0   | 2133 | 1.9842          |
+| 0.5648        | 10.0  | 2370 | 2.0032          |
 
 
 ### Framework versions

all_results.json

@@ -1,14 +1,9 @@
 {
-    "epoch": 4.988913525498892,
-    "eval_loss": 1.8087279796600342,
-    "eval_runtime": 0.3834,
-    "eval_samples": 16,
-    "eval_samples_per_second": 2.608,
-    "eval_steps_per_second": 2.608,
-    "total_flos": 1.6629843858229821e+18,
-    "train_loss": 1.390608725865682,
-    "train_runtime": 3610.9267,
+    "epoch": 10.0,
+    "total_flos": 3.5097090775444357e+18,
+    "train_loss": 0.7390849222110797,
+    "train_runtime": 8188.9555,
     "train_samples": 92634,
-    "train_samples_per_second": 9.977,
-    "train_steps_per_second": 0.312
+    "train_samples_per_second": 9.243,
+    "train_steps_per_second": 0.289
 }

train_results.json

@@ -1,9 +1,9 @@
 {
-    "epoch": 4.988913525498892,
-    "total_flos": 1.6629843858229821e+18,
-    "train_loss": 1.390608725865682,
-    "train_runtime": 3610.9267,
+    "epoch": 10.0,
+    "total_flos": 3.5097090775444357e+18,
+    "train_loss": 0.7390849222110797,
+    "train_runtime": 8188.9555,
     "train_samples": 92634,
-    "train_samples_per_second": 9.977,
-    "train_steps_per_second": 0.312
+    "train_samples_per_second": 9.243,
+    "train_steps_per_second": 0.289
 }