End of training

README.md

@@ -1,14 +1,14 @@
 ---
 library_name: peft
 license: apache-2.0
-base_model: llamafactory/tiny-random-Llama-3
+base_model: unsloth/Qwen2.5-Coder-1.5B-Instruct
 tags:
 - axolotl
 - generated_from_trainer
 datasets:
-- Paladiso/dataset_8c492e8c-0607-4470-a6df-b4d669ca4790
+- Paladiso/dataset_5ded376d-67f1-49d0-ab3d-1b1b59aa00cb
 model-index:
-- name: 2d119c47-07ab-425d-bf69-5940cfef50a9
+- name: c89227a8-2bbd-4dc1-93e5-c55c70367a30
   results: []
 ---
 
@@ -21,18 +21,18 @@ should probably proofread and complete it, then remove this comment. -->
 axolotl version: `0.6.0`
 ```yaml
 adapter: lora
-base_model: llamafactory/tiny-random-Llama-3
+base_model: unsloth/Qwen2.5-Coder-1.5B-Instruct
 bf16: auto
 chat_template: llama3
 dataset_prepared_path: /workspace/axolotl/data/prepared
 datasets:
 - ds_type: json
   format: custom
-  path: Paladiso/dataset_8c492e8c-0607-4470-a6df-b4d669ca4790
+  path: Paladiso/dataset_5ded376d-67f1-49d0-ab3d-1b1b59aa00cb
   type:
-    field_input: chosen
-    field_instruction: prompt
-    field_output: response
+    field_input: persona_ms
+    field_instruction: persona
+    field_output: context
     system_format: '{system}'
     system_prompt: ''
 debug: null
@@ -48,7 +48,7 @@ fsdp_config: null
 gradient_accumulation_steps: 4
 gradient_checkpointing: false
 group_by_length: false
-hub_model_id: Paladiso/2d119c47-07ab-425d-bf69-5940cfef50a9
+hub_model_id: Paladiso/c89227a8-2bbd-4dc1-93e5-c55c70367a30
 hub_private_repo: true
 hub_repo: null
 hub_strategy: checkpoint
@@ -79,8 +79,6 @@ sample_packing: false
 save_safetensors: true
 saves_per_epoch: 4
 sequence_len: 512
-special_tokens:
-  pad_token: <|eot_id|>
 strict: false
 tf32: false
 tokenizer_type: AutoTokenizer
@@ -90,10 +88,10 @@ use_accelerate: true
 val_set_size: 0.05
 wandb_entity: null
 wandb_mode: online
-wandb_name: 8c492e8c-0607-4470-a6df-b4d669ca4790
+wandb_name: 5ded376d-67f1-49d0-ab3d-1b1b59aa00cb
 wandb_project: Gradients-On-Demand
 wandb_run: your_name
-wandb_runid: 8c492e8c-0607-4470-a6df-b4d669ca4790
+wandb_runid: 5ded376d-67f1-49d0-ab3d-1b1b59aa00cb
 warmup_steps: 10
 weight_decay: 0.0
 xformers_attention: null
@@ -102,11 +100,11 @@ xformers_attention: null
 
 </details><br>
 
-# 2d119c47-07ab-425d-bf69-5940cfef50a9
+# c89227a8-2bbd-4dc1-93e5-c55c70367a30
 
-This model is a fine-tuned version of [llamafactory/tiny-random-Llama-3](https://huggingface.co/llamafactory/tiny-random-Llama-3) on the Paladiso/dataset_8c492e8c-0607-4470-a6df-b4d669ca4790 dataset.
+This model is a fine-tuned version of [unsloth/Qwen2.5-Coder-1.5B-Instruct](https://huggingface.co/unsloth/Qwen2.5-Coder-1.5B-Instruct) on the Paladiso/dataset_5ded376d-67f1-49d0-ab3d-1b1b59aa00cb dataset.
 It achieves the following results on the evaluation set:
-- Loss: 11.7661
+- Loss: 2.7890
 
 ## Model description
 
@@ -140,9 +138,9 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch  | Step | Validation Loss |
 |:-------------:|:------:|:----:|:---------------:|
-| 11.7623       | 0.0002 | 3    | 11.7663         |
-| 11.7701       | 0.0004 | 6    | 11.7662         |
-| 11.7674       | 0.0006 | 9    | 11.7661         |
+| 2.968         | 0.0001 | 3    | 2.9008          |
+| 3.1394        | 0.0003 | 6    | 2.8616          |
+| 3.0318        | 0.0004 | 9    | 2.7890          |
 
 
 ### Framework versions