End of training

README.md

@@ -1,14 +1,13 @@
 ---
 library_name: peft
-license: apache-2.0
-base_model: llamafactory/tiny-random-Llama-3
+base_model: Korabbit/llama-2-ko-7b
 tags:
 - axolotl
 - generated_from_trainer
 datasets:
-- Aivesa/dataset_3196ce3b-6aa4-47f9-a5b7-bfd4b34e4b38
+- Aivesa/dataset_ea212b2c-7a64-40e6-b0ba-090dcd4d4cf9
 model-index:
-- name: 239d5457-8520-43fb-97c4-b96a2ac31210
+- name: 7f64d457-d879-428f-895a-d47c066eebc8
   results: []
 ---
 
@@ -21,17 +20,18 @@ 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: Korabbit/llama-2-ko-7b
 bf16: auto
 chat_template: llama3
 dataset_prepared_path: /workspace/axolotl/data/prepared
 datasets:
 - ds_type: json
   format: custom
-  path: Aivesa/dataset_3196ce3b-6aa4-47f9-a5b7-bfd4b34e4b38
+  path: Aivesa/dataset_ea212b2c-7a64-40e6-b0ba-090dcd4d4cf9
   type:
-    field_instruction: inputs
-    field_output: targets
+    field_input: input
+    field_instruction: instruction
+    field_output: output
     system_format: '{system}'
     system_prompt: ''
 debug: null
@@ -47,7 +47,7 @@ fsdp_config: null
 gradient_accumulation_steps: 4
 gradient_checkpointing: false
 group_by_length: false
-hub_model_id: Aivesa/239d5457-8520-43fb-97c4-b96a2ac31210
+hub_model_id: Aivesa/7f64d457-d879-428f-895a-d47c066eebc8
 hub_private_repo: true
 hub_repo: null
 hub_strategy: checkpoint
@@ -79,7 +79,7 @@ save_safetensors: true
 saves_per_epoch: 4
 sequence_len: 512
 special_tokens:
-  pad_token: <|eot_id|>
+  pad_token: </s>
 strict: false
 tf32: false
 tokenizer_type: AutoTokenizer
@@ -89,10 +89,10 @@ use_accelerate: true
 val_set_size: 0.05
 wandb_entity: null
 wandb_mode: online
-wandb_name: 3196ce3b-6aa4-47f9-a5b7-bfd4b34e4b38
+wandb_name: ea212b2c-7a64-40e6-b0ba-090dcd4d4cf9
 wandb_project: Gradients-On-Demand
 wandb_run: your_name
-wandb_runid: 3196ce3b-6aa4-47f9-a5b7-bfd4b34e4b38
+wandb_runid: ea212b2c-7a64-40e6-b0ba-090dcd4d4cf9
 warmup_steps: 10
 weight_decay: 0.0
 xformers_attention: null
@@ -101,11 +101,11 @@ xformers_attention: null
 
 </details><br>
 
-# 239d5457-8520-43fb-97c4-b96a2ac31210
+# 7f64d457-d879-428f-895a-d47c066eebc8
 
-This model is a fine-tuned version of [llamafactory/tiny-random-Llama-3](https://huggingface.co/llamafactory/tiny-random-Llama-3) on the Aivesa/dataset_3196ce3b-6aa4-47f9-a5b7-bfd4b34e4b38 dataset.
+This model is a fine-tuned version of [Korabbit/llama-2-ko-7b](https://huggingface.co/Korabbit/llama-2-ko-7b) on the Aivesa/dataset_ea212b2c-7a64-40e6-b0ba-090dcd4d4cf9 dataset.
 It achieves the following results on the evaluation set:
-- Loss: 11.7598
+- Loss: 1.1518
 
 ## Model description
 
@@ -139,9 +139,9 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch  | Step | Validation Loss |
 |:-------------:|:------:|:----:|:---------------:|
-| 11.7633       | 0.0054 | 3    | 11.7599         |
-| 11.7713       | 0.0107 | 6    | 11.7599         |
-| 11.76         | 0.0161 | 9    | 11.7598         |
+| 3.2394        | 0.0108 | 3    | 3.6426          |
+| 2.9842        | 0.0215 | 6    | 2.8471          |
+| 1.676         | 0.0323 | 9    | 1.1518          |
 
 
 ### Framework versions