End of training

README.md

@@ -1,14 +1,14 @@
 ---
 library_name: peft
 license: apache-2.0
-base_model: Qwen/Qwen2.5-Math-7B-Instruct
+base_model: unsloth/Qwen2.5-0.5B-Instruct
 tags:
 - axolotl
 - generated_from_trainer
 datasets:
-- Paladiso/dataset_8d0a43d1-c494-4713-88d3-33e90ed16b11
+- Paladiso/dataset_e61a2e67-2ef6-4c56-bc3d-5981b6729822
 model-index:
-- name: 5ac2d2e1-dd81-4f3e-91df-8736e18a686a
+- name: 0d028e06-1aab-4890-b182-1143f59e3b1a
   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: Qwen/Qwen2.5-Math-7B-Instruct
+base_model: unsloth/Qwen2.5-0.5B-Instruct
 bf16: auto
 chat_template: llama3
 dataset_prepared_path: /workspace/axolotl/data/prepared
 datasets:
 - ds_type: json
   format: custom
-  path: Paladiso/dataset_8d0a43d1-c494-4713-88d3-33e90ed16b11
+  path: Paladiso/dataset_e61a2e67-2ef6-4c56-bc3d-5981b6729822
   type:
-    field_instruction: text_original
-    field_output: text_normalized
+    field_input: SMILES
+    field_instruction: Name
+    field_output: Descriptor
     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: Paladiso/5ac2d2e1-dd81-4f3e-91df-8736e18a686a
+hub_model_id: Paladiso/0d028e06-1aab-4890-b182-1143f59e3b1a
 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: 8d0a43d1-c494-4713-88d3-33e90ed16b11
+wandb_name: e61a2e67-2ef6-4c56-bc3d-5981b6729822
 wandb_project: Gradients-On-Demand
 wandb_run: your_name
-wandb_runid: 8d0a43d1-c494-4713-88d3-33e90ed16b11
+wandb_runid: e61a2e67-2ef6-4c56-bc3d-5981b6729822
 warmup_steps: 10
 weight_decay: 0.0
 xformers_attention: null
@@ -99,11 +100,11 @@ xformers_attention: null
 
 </details><br>
 
-# 5ac2d2e1-dd81-4f3e-91df-8736e18a686a
+# 0d028e06-1aab-4890-b182-1143f59e3b1a
 
-This model is a fine-tuned version of [Qwen/Qwen2.5-Math-7B-Instruct](https://huggingface.co/Qwen/Qwen2.5-Math-7B-Instruct) on the Paladiso/dataset_8d0a43d1-c494-4713-88d3-33e90ed16b11 dataset.
+This model is a fine-tuned version of [unsloth/Qwen2.5-0.5B-Instruct](https://huggingface.co/unsloth/Qwen2.5-0.5B-Instruct) on the Paladiso/dataset_e61a2e67-2ef6-4c56-bc3d-5981b6729822 dataset.
 It achieves the following results on the evaluation set:
-- Loss: 2.1437
+- Loss: 2.6470
 
 ## Model description
 
@@ -137,9 +138,9 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch  | Step | Validation Loss |
 |:-------------:|:------:|:----:|:---------------:|
-| 2.203         | 0.0001 | 3    | 2.8900          |
-| 2.5232        | 0.0001 | 6    | 2.7809          |
-| 3.5169        | 0.0002 | 9    | 2.1437          |
+| 2.4995        | 0.0019 | 3    | 2.8456          |
+| 2.8845        | 0.0037 | 6    | 2.7812          |
+| 2.7788        | 0.0056 | 9    | 2.6470          |
 
 
 ### Framework versions