End of training

README.md

@@ -1,14 +1,14 @@
 ---
 library_name: peft
 license: apache-2.0
-base_model: unsloth/Qwen2-0.5B-Instruct
+base_model: EleutherAI/pythia-1b
 tags:
 - axolotl
 - generated_from_trainer
 datasets:
-- Aivesa/dataset_fdd2902e-b68a-478a-8c73-81802471323d
+- Aivesa/dataset_461352e1-3da4-46a2-a77a-265bfac04700
 model-index:
-- name: 38f5cfa4-d2db-44cd-9dae-55988102a590
+- name: 6dd4d67e-276b-4a84-a5c8-a7851742b7b4
   results: []
 ---
 
@@ -21,17 +21,17 @@ should probably proofread and complete it, then remove this comment. -->
 axolotl version: `0.6.0`
 ```yaml
 adapter: lora
-base_model: unsloth/Qwen2-0.5B-Instruct
+base_model: EleutherAI/pythia-1b
 bf16: auto
 chat_template: llama3
 dataset_prepared_path: /workspace/axolotl/data/prepared
 datasets:
 - ds_type: json
   format: custom
-  path: Aivesa/dataset_fdd2902e-b68a-478a-8c73-81802471323d
+  path: Aivesa/dataset_461352e1-3da4-46a2-a77a-265bfac04700
   type:
     field_instruction: prompt
-    field_output: chosen
+    field_output: label
     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/38f5cfa4-d2db-44cd-9dae-55988102a590
+hub_model_id: Aivesa/6dd4d67e-276b-4a84-a5c8-a7851742b7b4
 hub_private_repo: true
 hub_repo: null
 hub_strategy: checkpoint
@@ -78,6 +78,8 @@ sample_packing: false
 save_safetensors: true
 saves_per_epoch: 4
 sequence_len: 512
+special_tokens:
+  pad_token: <|endoftext|>
 strict: false
 tf32: false
 tokenizer_type: AutoTokenizer
@@ -87,10 +89,10 @@ use_accelerate: true
 val_set_size: 0.05
 wandb_entity: null
 wandb_mode: online
-wandb_name: fdd2902e-b68a-478a-8c73-81802471323d
+wandb_name: 461352e1-3da4-46a2-a77a-265bfac04700
 wandb_project: Gradients-On-Demand
 wandb_run: your_name
-wandb_runid: fdd2902e-b68a-478a-8c73-81802471323d
+wandb_runid: 461352e1-3da4-46a2-a77a-265bfac04700
 warmup_steps: 10
 weight_decay: 0.0
 xformers_attention: null
@@ -99,11 +101,11 @@ xformers_attention: null
 
 </details><br>
 
-# 38f5cfa4-d2db-44cd-9dae-55988102a590
+# 6dd4d67e-276b-4a84-a5c8-a7851742b7b4
 
-This model is a fine-tuned version of [unsloth/Qwen2-0.5B-Instruct](https://huggingface.co/unsloth/Qwen2-0.5B-Instruct) on the Aivesa/dataset_fdd2902e-b68a-478a-8c73-81802471323d dataset.
+This model is a fine-tuned version of [EleutherAI/pythia-1b](https://huggingface.co/EleutherAI/pythia-1b) on the Aivesa/dataset_461352e1-3da4-46a2-a77a-265bfac04700 dataset.
 It achieves the following results on the evaluation set:
-- Loss: 2.6926
+- Loss: 2.1589
 
 ## Model description
 
@@ -137,9 +139,9 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch  | Step | Validation Loss |
 |:-------------:|:------:|:----:|:---------------:|
-| 2.7649        | 0.0042 | 3    | 3.1153          |
-| 3.1474        | 0.0083 | 6    | 2.9960          |
-| 2.6516        | 0.0125 | 9    | 2.6926          |
+| 9.3476        | 0.0002 | 3    | 2.3903          |
+| 10.2456       | 0.0003 | 6    | 2.3352          |
+| 8.507         | 0.0005 | 9    | 2.1589          |
 
 
 ### Framework versions