End of training

README.md

@@ -1,13 +1,14 @@
 ---
 library_name: peft
-base_model: unsloth/Hermes-3-Llama-3.1-8B
+license: apache-2.0
+base_model: unsloth/Qwen2.5-1.5B
 tags:
 - axolotl
 - generated_from_trainer
 datasets:
-- Aivesa/dataset_5febd7f3-e2ea-40a1-b878-8e9354f2a1e3
+- Aivesa/dataset_7fc40eed-6bc5-4668-b817-a908f0a659fe
 model-index:
-- name: dcb60e49-28f7-4991-a4cc-b9ed2381c40b
+- name: fc7797c5-8beb-45b4-9e9a-bb988d029b44
   results: []
 ---
 
@@ -20,17 +21,17 @@ should probably proofread and complete it, then remove this comment. -->
 axolotl version: `0.6.0`
 ```yaml
 adapter: lora
-base_model: unsloth/Hermes-3-Llama-3.1-8B
+base_model: unsloth/Qwen2.5-1.5B
 bf16: auto
 chat_template: llama3
 dataset_prepared_path: /workspace/axolotl/data/prepared
 datasets:
 - ds_type: json
   format: custom
-  path: Aivesa/dataset_5febd7f3-e2ea-40a1-b878-8e9354f2a1e3
+  path: Aivesa/dataset_7fc40eed-6bc5-4668-b817-a908f0a659fe
   type:
-    field_instruction: SeedRule
-    field_output: prompt
+    field_instruction: abstract
+    field_output: title
     system_format: '{system}'
     system_prompt: ''
 debug: null
@@ -46,7 +47,7 @@ fsdp_config: null
 gradient_accumulation_steps: 4
 gradient_checkpointing: false
 group_by_length: false
-hub_model_id: Aivesa/dcb60e49-28f7-4991-a4cc-b9ed2381c40b
+hub_model_id: Aivesa/fc7797c5-8beb-45b4-9e9a-bb988d029b44
 hub_private_repo: true
 hub_repo: null
 hub_strategy: checkpoint
@@ -86,10 +87,10 @@ use_accelerate: true
 val_set_size: 0.05
 wandb_entity: null
 wandb_mode: online
-wandb_name: 5febd7f3-e2ea-40a1-b878-8e9354f2a1e3
+wandb_name: 7fc40eed-6bc5-4668-b817-a908f0a659fe
 wandb_project: Gradients-On-Demand
 wandb_run: your_name
-wandb_runid: 5febd7f3-e2ea-40a1-b878-8e9354f2a1e3
+wandb_runid: 7fc40eed-6bc5-4668-b817-a908f0a659fe
 warmup_steps: 10
 weight_decay: 0.0
 xformers_attention: null
@@ -98,11 +99,11 @@ xformers_attention: null
 
 </details><br>
 
-# dcb60e49-28f7-4991-a4cc-b9ed2381c40b
+# fc7797c5-8beb-45b4-9e9a-bb988d029b44
 
-This model is a fine-tuned version of [unsloth/Hermes-3-Llama-3.1-8B](https://huggingface.co/unsloth/Hermes-3-Llama-3.1-8B) on the Aivesa/dataset_5febd7f3-e2ea-40a1-b878-8e9354f2a1e3 dataset.
+This model is a fine-tuned version of [unsloth/Qwen2.5-1.5B](https://huggingface.co/unsloth/Qwen2.5-1.5B) on the Aivesa/dataset_7fc40eed-6bc5-4668-b817-a908f0a659fe dataset.
 It achieves the following results on the evaluation set:
-- Loss: 2.8035
+- Loss: 2.3698
 
 ## Model description
 
@@ -136,9 +137,9 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch  | Step | Validation Loss |
 |:-------------:|:------:|:----:|:---------------:|
-| 2.7241        | 0.0038 | 3    | 3.1756          |
-| 2.4554        | 0.0076 | 6    | 3.1208          |
-| 2.9521        | 0.0115 | 9    | 2.8035          |
+| 2.3275        | 0.0004 | 3    | 2.8304          |
+| 2.4043        | 0.0007 | 6    | 2.7153          |
+| 2.856         | 0.0011 | 9    | 2.3698          |
 
 
 ### Framework versions