End of training

README.md

@@ -1,14 +1,14 @@
 ---
 library_name: peft
-license: llama3
-base_model: unsloth/llama-3-8b-Instruct
+license: other
+base_model: facebook/opt-1.3b
 tags:
 - axolotl
 - generated_from_trainer
 datasets:
-- Paladiso/dataset_695de20c-0af8-4b07-94bc-5ccdfcc25776
+- Paladiso/dataset_e61ef3ef-654f-42b4-9405-ed9d0cb7ec9e
 model-index:
-- name: a1ea3ebd-561e-45da-86b1-ff6386e13625
+- name: 38df3cea-ad5b-496c-9e37-bab21c6decfc
   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: unsloth/llama-3-8b-Instruct
+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_695de20c-0af8-4b07-94bc-5ccdfcc25776
+  path: Paladiso/dataset_e61ef3ef-654f-42b4-9405-ed9d0cb7ec9e
   type:
-    field_input: parent_id
-    field_instruction: role
-    field_output: text
+    field_instruction: question
+    field_output: best
     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/a1ea3ebd-561e-45da-86b1-ff6386e13625
+hub_model_id: Paladiso/38df3cea-ad5b-496c-9e37-bab21c6decfc
 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: 695de20c-0af8-4b07-94bc-5ccdfcc25776
+wandb_name: e61ef3ef-654f-42b4-9405-ed9d0cb7ec9e
 wandb_project: Gradients-On-Demand
 wandb_run: your_name
-wandb_runid: 695de20c-0af8-4b07-94bc-5ccdfcc25776
+wandb_runid: e61ef3ef-654f-42b4-9405-ed9d0cb7ec9e
 warmup_steps: 10
 weight_decay: 0.0
 xformers_attention: null
@@ -100,11 +99,11 @@ xformers_attention: null
 
 </details><br>
 
-# a1ea3ebd-561e-45da-86b1-ff6386e13625
+# 38df3cea-ad5b-496c-9e37-bab21c6decfc
 
-This model is a fine-tuned version of [unsloth/llama-3-8b-Instruct](https://huggingface.co/unsloth/llama-3-8b-Instruct) on the Paladiso/dataset_695de20c-0af8-4b07-94bc-5ccdfcc25776 dataset.
+This model is a fine-tuned version of [facebook/opt-1.3b](https://huggingface.co/facebook/opt-1.3b) on the Paladiso/dataset_e61ef3ef-654f-42b4-9405-ed9d0cb7ec9e dataset.
 It achieves the following results on the evaluation set:
-- Loss: 2.0698
+- Loss: 1.4838
 
 ## Model description
 
@@ -138,9 +137,9 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch  | Step | Validation Loss |
 |:-------------:|:------:|:----:|:---------------:|
-| 2.4536        | 0.0011 | 3    | 2.3868          |
-| 2.1539        | 0.0021 | 6    | 2.3161          |
-| 1.9608        | 0.0032 | 9    | 2.0698          |
+| 6.1042        | 0.0002 | 3    | 1.5160          |
+| 6.4478        | 0.0005 | 6    | 1.5023          |
+| 5.9355        | 0.0007 | 9    | 1.4838          |
 
 
 ### Framework versions