End of training

README.md

@@ -1,14 +1,14 @@
 ---
 library_name: peft
-license: apache-2.0
-base_model: NousResearch/Nous-Hermes-2-Mistral-7B-DPO
+license: llama3
+base_model: Orenguteng/Llama-3-8B-Lexi-Uncensored
 tags:
 - axolotl
 - generated_from_trainer
 datasets:
-- Paladiso/dataset_10f0655a-a09c-4504-a15e-0903bd3b37a1
+- Paladiso/dataset_dc96f1c3-2ce8-41ea-ad70-8bfc8b7c58e7
 model-index:
-- name: 6ed5b447-d630-4814-a9b0-e5021ca0f2bc
+- name: e307da6f-722d-4008-b2b6-3ae184f063ee
   results: []
 ---
 
@@ -21,18 +21,17 @@ should probably proofread and complete it, then remove this comment. -->
 axolotl version: `0.6.0`
 ```yaml
 adapter: lora
-base_model: NousResearch/Nous-Hermes-2-Mistral-7B-DPO
+base_model: Orenguteng/Llama-3-8B-Lexi-Uncensored
 bf16: auto
 chat_template: llama3
 dataset_prepared_path: /workspace/axolotl/data/prepared
 datasets:
 - ds_type: json
   format: custom
-  path: Paladiso/dataset_10f0655a-a09c-4504-a15e-0903bd3b37a1
+  path: Paladiso/dataset_dc96f1c3-2ce8-41ea-ad70-8bfc8b7c58e7
   type:
-    field_input: system_prompt
-    field_instruction: question
-    field_output: response
+    field_instruction: prompt
+    field_output: label
     system_format: '{system}'
     system_prompt: ''
 debug: null
@@ -48,7 +47,7 @@ fsdp_config: null
 gradient_accumulation_steps: 4
 gradient_checkpointing: false
 group_by_length: false
-hub_model_id: Paladiso/6ed5b447-d630-4814-a9b0-e5021ca0f2bc
+hub_model_id: Paladiso/e307da6f-722d-4008-b2b6-3ae184f063ee
 hub_private_repo: true
 hub_repo: null
 hub_strategy: checkpoint
@@ -88,10 +87,10 @@ use_accelerate: true
 val_set_size: 0.05
 wandb_entity: null
 wandb_mode: online
-wandb_name: 10f0655a-a09c-4504-a15e-0903bd3b37a1
+wandb_name: dc96f1c3-2ce8-41ea-ad70-8bfc8b7c58e7
 wandb_project: Gradients-On-Demand
 wandb_run: your_name
-wandb_runid: 10f0655a-a09c-4504-a15e-0903bd3b37a1
+wandb_runid: dc96f1c3-2ce8-41ea-ad70-8bfc8b7c58e7
 warmup_steps: 10
 weight_decay: 0.0
 xformers_attention: null
@@ -100,11 +99,11 @@ xformers_attention: null
 
 </details><br>
 
-# 6ed5b447-d630-4814-a9b0-e5021ca0f2bc
+# e307da6f-722d-4008-b2b6-3ae184f063ee
 
-This model is a fine-tuned version of [NousResearch/Nous-Hermes-2-Mistral-7B-DPO](https://huggingface.co/NousResearch/Nous-Hermes-2-Mistral-7B-DPO) on the Paladiso/dataset_10f0655a-a09c-4504-a15e-0903bd3b37a1 dataset.
+This model is a fine-tuned version of [Orenguteng/Llama-3-8B-Lexi-Uncensored](https://huggingface.co/Orenguteng/Llama-3-8B-Lexi-Uncensored) on the Paladiso/dataset_dc96f1c3-2ce8-41ea-ad70-8bfc8b7c58e7 dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.5881
+- Loss: 2.0576
 
 ## Model description
 
@@ -138,9 +137,9 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch  | Step | Validation Loss |
 |:-------------:|:------:|:----:|:---------------:|
-| 3.2984        | 0.0002 | 3    | 0.7134          |
-| 2.956         | 0.0005 | 6    | 0.6448          |
-| 2.4023        | 0.0007 | 9    | 0.5881          |
+| 2.3316        | 0.0002 | 3    | 2.4105          |
+| 2.3702        | 0.0003 | 6    | 2.2689          |
+| 2.1011        | 0.0005 | 9    | 2.0576          |
 
 
 ### Framework versions