End of training

README.md

@@ -1,13 +1,14 @@
 ---
 library_name: peft
-base_model: lmsys/vicuna-7b-v1.3
+license: llama3.2
+base_model: NousResearch/Llama-3.2-1B
 tags:
 - axolotl
 - generated_from_trainer
 datasets:
-- Paladiso/dataset_6f5a9d02-d1dd-4210-bbe2-5693802a786a
+- Paladiso/dataset_92a8bc35-2c7a-4a62-8dca-171a04f7b68a
 model-index:
-- name: 0643945e-0ddd-4b59-9c63-173e3ee2fe70
+- name: d340e4a1-0db1-4576-ba99-7226de7d1422
   results: []
 ---
 
@@ -20,18 +21,18 @@ should probably proofread and complete it, then remove this comment. -->
 axolotl version: `0.6.0`
 ```yaml
 adapter: lora
-base_model: lmsys/vicuna-7b-v1.3
+base_model: NousResearch/Llama-3.2-1B
 bf16: auto
 chat_template: llama3
 dataset_prepared_path: /workspace/axolotl/data/prepared
 datasets:
 - ds_type: json
   format: custom
-  path: Paladiso/dataset_6f5a9d02-d1dd-4210-bbe2-5693802a786a
+  path: Paladiso/dataset_92a8bc35-2c7a-4a62-8dca-171a04f7b68a
   type:
-    field_input: ''
-    field_instruction: problem
-    field_output: solution
+    field_input: activity
+    field_instruction: topic
+    field_output: text
     system_format: '{system}'
     system_prompt: ''
 debug: null
@@ -47,7 +48,7 @@ fsdp_config: null
 gradient_accumulation_steps: 4
 gradient_checkpointing: false
 group_by_length: false
-hub_model_id: Paladiso/0643945e-0ddd-4b59-9c63-173e3ee2fe70
+hub_model_id: Paladiso/d340e4a1-0db1-4576-ba99-7226de7d1422
 hub_private_repo: true
 hub_repo: null
 hub_strategy: checkpoint
@@ -78,6 +79,8 @@ sample_packing: false
 save_safetensors: true
 saves_per_epoch: 4
 sequence_len: 512
+special_tokens:
+  pad_token: <|end_of_text|>
 strict: false
 tf32: false
 tokenizer_type: AutoTokenizer
@@ -87,10 +90,10 @@ use_accelerate: true
 val_set_size: 0.05
 wandb_entity: null
 wandb_mode: online
-wandb_name: 6f5a9d02-d1dd-4210-bbe2-5693802a786a
+wandb_name: 92a8bc35-2c7a-4a62-8dca-171a04f7b68a
 wandb_project: Gradients-On-Demand
 wandb_run: your_name
-wandb_runid: 6f5a9d02-d1dd-4210-bbe2-5693802a786a
+wandb_runid: 92a8bc35-2c7a-4a62-8dca-171a04f7b68a
 warmup_steps: 10
 weight_decay: 0.0
 xformers_attention: null
@@ -99,11 +102,11 @@ xformers_attention: null
 
 </details><br>
 
-# 0643945e-0ddd-4b59-9c63-173e3ee2fe70
+# d340e4a1-0db1-4576-ba99-7226de7d1422
 
-This model is a fine-tuned version of [lmsys/vicuna-7b-v1.3](https://huggingface.co/lmsys/vicuna-7b-v1.3) on the Paladiso/dataset_6f5a9d02-d1dd-4210-bbe2-5693802a786a dataset.
+This model is a fine-tuned version of [NousResearch/Llama-3.2-1B](https://huggingface.co/NousResearch/Llama-3.2-1B) on the Paladiso/dataset_92a8bc35-2c7a-4a62-8dca-171a04f7b68a dataset.
 It achieves the following results on the evaluation set:
-- Loss: 1.0326
+- Loss: 2.0209
 
 ## Model description
 
@@ -137,9 +140,9 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch  | Step | Validation Loss |
 |:-------------:|:------:|:----:|:---------------:|
-| 1.3042        | 0.0009 | 3    | 1.3808          |
-| 1.1691        | 0.0018 | 6    | 1.3009          |
-| 1.0312        | 0.0027 | 9    | 1.0326          |
+| 2.2288        | 0.0014 | 3    | 2.2458          |
+| 2.1137        | 0.0028 | 6    | 2.1610          |
+| 2.0657        | 0.0042 | 9    | 2.0209          |
 
 
 ### Framework versions