End of training

README.md

@@ -1,14 +1,14 @@
 ---
 library_name: peft
 license: other
-base_model: facebook/opt-125m
+base_model: facebook/opt-1.3b
 tags:
 - axolotl
 - generated_from_trainer
 datasets:
-- Paladiso/dataset_23fa5d51-8fcd-4655-8190-c4e501fc6353
+- Paladiso/dataset_46dd44b3-e4e4-4c44-8605-e8e8b6dd956e
 model-index:
-- name: bcfa40dc-a9c9-4105-b999-e15b08e40d79
+- name: d98dba11-2239-4058-9db7-c5ed3595c1b9
   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: facebook/opt-125m
+base_model: facebook/opt-1.3b
 bf16: auto
 chat_template: llama3
 dataset_prepared_path: /workspace/axolotl/data/prepared
 datasets:
 - ds_type: json
   format: custom
-  path: Paladiso/dataset_23fa5d51-8fcd-4655-8190-c4e501fc6353
+  path: Paladiso/dataset_46dd44b3-e4e4-4c44-8605-e8e8b6dd956e
   type:
-    field_instruction: prompt
-    field_output: response_a
+    field_input: context
+    field_instruction: question
+    field_output: answers
     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/bcfa40dc-a9c9-4105-b999-e15b08e40d79
+hub_model_id: Paladiso/d98dba11-2239-4058-9db7-c5ed3595c1b9
 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: 23fa5d51-8fcd-4655-8190-c4e501fc6353
+wandb_name: 46dd44b3-e4e4-4c44-8605-e8e8b6dd956e
 wandb_project: Gradients-On-Demand
 wandb_run: your_name
-wandb_runid: 23fa5d51-8fcd-4655-8190-c4e501fc6353
+wandb_runid: 46dd44b3-e4e4-4c44-8605-e8e8b6dd956e
 warmup_steps: 10
 weight_decay: 0.0
 xformers_attention: null
@@ -99,11 +100,11 @@ xformers_attention: null
 
 </details><br>
 
-# bcfa40dc-a9c9-4105-b999-e15b08e40d79
+# d98dba11-2239-4058-9db7-c5ed3595c1b9
 
-This model is a fine-tuned version of [facebook/opt-125m](https://huggingface.co/facebook/opt-125m) on the Paladiso/dataset_23fa5d51-8fcd-4655-8190-c4e501fc6353 dataset.
+This model is a fine-tuned version of [facebook/opt-1.3b](https://huggingface.co/facebook/opt-1.3b) on the Paladiso/dataset_46dd44b3-e4e4-4c44-8605-e8e8b6dd956e dataset.
 It achieves the following results on the evaluation set:
-- Loss: 2.3607
+- Loss: 2.6615
 
 ## Model description
 
@@ -137,9 +138,9 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch  | Step | Validation Loss |
 |:-------------:|:------:|:----:|:---------------:|
-| 10.0335       | 0.0004 | 3    | 2.4104          |
-| 8.8786        | 0.0009 | 6    | 2.3934          |
-| 10.1423       | 0.0013 | 9    | 2.3607          |
+| 15.8207       | 0.0003 | 3    | 4.1120          |
+| 17.5493       | 0.0005 | 6    | 3.6225          |
+| 11.6362       | 0.0008 | 9    | 2.6615          |
 
 
 ### Framework versions