End of training

README.md

@@ -1,13 +1,13 @@
 ---
 library_name: peft
-base_model: katuni4ka/tiny-random-qwen1.5-moe
+base_model: Korabbit/llama-2-ko-7b
 tags:
 - axolotl
 - generated_from_trainer
 datasets:
-- Paladiso/dataset_5e3967ba-d17c-4d93-91ec-a23620abb5dc
+- Paladiso/dataset_6f876187-6d0a-4410-b9c7-16623cec1fad
 model-index:
-- name: 52d5acd4-4d39-43e2-8519-fe17241200b6
+- name: bb558fe8-9262-4aef-9da8-8c8ad8a19526
   results: []
 ---
 
@@ -20,17 +20,18 @@ should probably proofread and complete it, then remove this comment. -->
 axolotl version: `0.6.0`
 ```yaml
 adapter: lora
-base_model: katuni4ka/tiny-random-qwen1.5-moe
+base_model: Korabbit/llama-2-ko-7b
 bf16: auto
 chat_template: llama3
 dataset_prepared_path: /workspace/axolotl/data/prepared
 datasets:
 - ds_type: json
   format: custom
-  path: Paladiso/dataset_5e3967ba-d17c-4d93-91ec-a23620abb5dc
+  path: Paladiso/dataset_6f876187-6d0a-4410-b9c7-16623cec1fad
   type:
-    field_instruction: instruction
-    field_output: output
+    field_input: Description
+    field_instruction: Prompt
+    field_output: GT
     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: Paladiso/52d5acd4-4d39-43e2-8519-fe17241200b6
+hub_model_id: Paladiso/bb558fe8-9262-4aef-9da8-8c8ad8a19526
 hub_private_repo: true
 hub_repo: null
 hub_strategy: checkpoint
@@ -77,6 +78,8 @@ sample_packing: false
 save_safetensors: true
 saves_per_epoch: 4
 sequence_len: 512
+special_tokens:
+  pad_token: </s>
 strict: false
 tf32: false
 tokenizer_type: AutoTokenizer
@@ -86,10 +89,10 @@ use_accelerate: true
 val_set_size: 0.05
 wandb_entity: null
 wandb_mode: online
-wandb_name: 5e3967ba-d17c-4d93-91ec-a23620abb5dc
+wandb_name: 6f876187-6d0a-4410-b9c7-16623cec1fad
 wandb_project: Gradients-On-Demand
 wandb_run: your_name
-wandb_runid: 5e3967ba-d17c-4d93-91ec-a23620abb5dc
+wandb_runid: 6f876187-6d0a-4410-b9c7-16623cec1fad
 warmup_steps: 10
 weight_decay: 0.0
 xformers_attention: null
@@ -98,11 +101,11 @@ xformers_attention: null
 
 </details><br>
 
-# 52d5acd4-4d39-43e2-8519-fe17241200b6
+# bb558fe8-9262-4aef-9da8-8c8ad8a19526
 
-This model is a fine-tuned version of [katuni4ka/tiny-random-qwen1.5-moe](https://huggingface.co/katuni4ka/tiny-random-qwen1.5-moe) on the Paladiso/dataset_5e3967ba-d17c-4d93-91ec-a23620abb5dc dataset.
+This model is a fine-tuned version of [Korabbit/llama-2-ko-7b](https://huggingface.co/Korabbit/llama-2-ko-7b) on the Paladiso/dataset_6f876187-6d0a-4410-b9c7-16623cec1fad dataset.
 It achieves the following results on the evaluation set:
-- Loss: 11.9366
+- Loss: 0.7600
 
 ## Model description
 
@@ -136,9 +139,9 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch  | Step | Validation Loss |
 |:-------------:|:------:|:----:|:---------------:|
-| 11.9333       | 0.0002 | 3    | 11.9376         |
-| 11.9358       | 0.0004 | 6    | 11.9372         |
-| 11.9335       | 0.0007 | 9    | 11.9366         |
+| 6.9483        | 0.0110 | 3    | 6.5820          |
+| 4.8242        | 0.0220 | 6    | 3.0552          |
+| 1.0649        | 0.0329 | 9    | 0.7600          |
 
 
 ### Framework versions