End of training

README.md

@@ -47,7 +47,7 @@ flash_attention: false
 fp16: true
 fsdp: null
 fsdp_config: null
-gradient_accumulation_steps: 8
+gradient_accumulation_steps: 24
 gradient_checkpointing: true
 group_by_length: false
 hub_model_id: error577/cb2f8889-7cc4-4506-8b3f-9bd75b18db3d
@@ -103,7 +103,7 @@ xformers_attention: null
 
 This model is a fine-tuned version of [NousResearch/Hermes-2-Pro-Llama-3-8B](https://huggingface.co/NousResearch/Hermes-2-Pro-Llama-3-8B) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 3.5717
+- Loss: 3.5350
 
 ## Model description
 
@@ -126,8 +126,8 @@ The following hyperparameters were used during training:
 - train_batch_size: 1
 - eval_batch_size: 1
 - seed: 42
-- gradient_accumulation_steps: 8
-- total_train_batch_size: 8
+- gradient_accumulation_steps: 24
+- total_train_batch_size: 24
 - optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
 - lr_scheduler_type: cosine
 - lr_scheduler_warmup_steps: 10
@@ -137,10 +137,10 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch  | Step | Validation Loss |
 |:-------------:|:------:|:----:|:---------------:|
-| 3.7463        | 0.0006 | 1    | 3.6378          |
-| 3.1519        | 0.0012 | 2    | 3.6369          |
-| 2.8665        | 0.0024 | 4    | 3.6247          |
-| 3.8022        | 0.0037 | 6    | 3.5717          |
+| 3.3147        | 0.0018 | 1    | 3.6378          |
+| 3.391         | 0.0037 | 2    | 3.6362          |
+| 3.105         | 0.0073 | 4    | 3.6184          |
+| 3.5426        | 0.0110 | 6    | 3.5350          |
 
 
 ### Framework versions

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4dffd45ec49ca61d4ac75475d1924718dc6512652ba53ce072798af922f2100b
+oid sha256:efdd3c2a88438634f09abfbbb9a9940eac8ca3674ad5002b0e66727e5b5e6b30
 size 42104138