End of training

README.md

@@ -1,14 +1,14 @@
 ---
 library_name: peft
-license: apache-2.0
-base_model: unsloth/Qwen2.5-Math-7B-Instruct
+license: mit
+base_model: NousResearch/Nous-Hermes-llama-2-7b
 tags:
 - axolotl
 - generated_from_trainer
 datasets:
-- Aivesa/dataset_e59ee620-6dc6-41b8-8080-617c8e470bc7
+- Aivesa/dataset_2a4738f3-e023-46f4-ae13-2b8b3faf7cb2
 model-index:
-- name: 77c11f2a-7fc5-42a9-af8d-7f0bb2c3679b
+- name: 37f4a9fa-b0b6-4fc9-b20b-00147fd7724e
   results: []
 ---
 
@@ -21,17 +21,18 @@ should probably proofread and complete it, then remove this comment. -->
 axolotl version: `0.6.0`
 ```yaml
 adapter: lora
-base_model: unsloth/Qwen2.5-Math-7B-Instruct
+base_model: NousResearch/Nous-Hermes-llama-2-7b
 bf16: auto
 chat_template: llama3
 dataset_prepared_path: /workspace/axolotl/data/prepared
 datasets:
 - ds_type: json
   format: custom
-  path: Aivesa/dataset_e59ee620-6dc6-41b8-8080-617c8e470bc7
+  path: Aivesa/dataset_2a4738f3-e023-46f4-ae13-2b8b3faf7cb2
   type:
-    field_instruction: instruction
-    field_output: output_1
+    field_input: title
+    field_instruction: category
+    field_output: title_category
     system_format: '{system}'
     system_prompt: ''
 debug: null
@@ -47,7 +48,7 @@ fsdp_config: null
 gradient_accumulation_steps: 4
 gradient_checkpointing: false
 group_by_length: false
-hub_model_id: Aivesa/77c11f2a-7fc5-42a9-af8d-7f0bb2c3679b
+hub_model_id: Aivesa/37f4a9fa-b0b6-4fc9-b20b-00147fd7724e
 hub_private_repo: true
 hub_repo: null
 hub_strategy: checkpoint
@@ -87,10 +88,10 @@ use_accelerate: true
 val_set_size: 0.05
 wandb_entity: null
 wandb_mode: online
-wandb_name: e59ee620-6dc6-41b8-8080-617c8e470bc7
+wandb_name: 2a4738f3-e023-46f4-ae13-2b8b3faf7cb2
 wandb_project: Gradients-On-Demand
 wandb_run: your_name
-wandb_runid: e59ee620-6dc6-41b8-8080-617c8e470bc7
+wandb_runid: 2a4738f3-e023-46f4-ae13-2b8b3faf7cb2
 warmup_steps: 10
 weight_decay: 0.0
 xformers_attention: null
@@ -99,11 +100,11 @@ xformers_attention: null
 
 </details><br>
 
-# 77c11f2a-7fc5-42a9-af8d-7f0bb2c3679b
+# 37f4a9fa-b0b6-4fc9-b20b-00147fd7724e
 
-This model is a fine-tuned version of [unsloth/Qwen2.5-Math-7B-Instruct](https://huggingface.co/unsloth/Qwen2.5-Math-7B-Instruct) on the Aivesa/dataset_e59ee620-6dc6-41b8-8080-617c8e470bc7 dataset.
+This model is a fine-tuned version of [NousResearch/Nous-Hermes-llama-2-7b](https://huggingface.co/NousResearch/Nous-Hermes-llama-2-7b) on the Aivesa/dataset_2a4738f3-e023-46f4-ae13-2b8b3faf7cb2 dataset.
 It achieves the following results on the evaluation set:
-- Loss: 2.6205
+- Loss: 0.4859
 
 ## Model description
 
@@ -137,9 +138,9 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch  | Step | Validation Loss |
 |:-------------:|:------:|:----:|:---------------:|
-| 2.4752        | 0.0015 | 3    | 2.6821          |
-| 2.6637        | 0.0031 | 6    | 2.6741          |
-| 2.5646        | 0.0046 | 9    | 2.6205          |
+| 1.5313        | 0.0000 | 3    | 1.4529          |
+| 1.3854        | 0.0001 | 6    | 1.1243          |
+| 0.8057        | 0.0001 | 9    | 0.4859          |
 
 
 ### Framework versions