End of training

README.md

@@ -1,13 +1,14 @@
 ---
 library_name: peft
-base_model: TitanML/tiny-mixtral
+license: apache-2.0
+base_model: EleutherAI/pythia-70m-deduped
 tags:
 - axolotl
 - generated_from_trainer
 datasets:
-- Aivesa/dataset_c45db162-924b-42af-bc3a-aa67e6972820
+- Aivesa/dataset_12e79e69-9871-48aa-a364-52587c3c39bc
 model-index:
-- name: 85a3b62a-09c1-44d6-af05-0f30aee76da5
+- name: df5493b9-6228-4cae-a15d-b3a2fd8fb6e3
   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: TitanML/tiny-mixtral
+base_model: EleutherAI/pythia-70m-deduped
 bf16: auto
 chat_template: llama3
 dataset_prepared_path: /workspace/axolotl/data/prepared
 datasets:
 - ds_type: json
   format: custom
-  path: Aivesa/dataset_c45db162-924b-42af-bc3a-aa67e6972820
+  path: Aivesa/dataset_12e79e69-9871-48aa-a364-52587c3c39bc
   type:
-    field_input: input
-    field_instruction: instruction
-    field_output: output
+    field_input: commit_message
+    field_instruction: func_before
+    field_output: func_after
     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: Aivesa/85a3b62a-09c1-44d6-af05-0f30aee76da5
+hub_model_id: Aivesa/df5493b9-6228-4cae-a15d-b3a2fd8fb6e3
 hub_private_repo: true
 hub_repo: null
 hub_strategy: checkpoint
@@ -79,7 +80,7 @@ save_safetensors: true
 saves_per_epoch: 4
 sequence_len: 512
 special_tokens:
-  pad_token: </s>
+  pad_token: <|endoftext|>
 strict: false
 tf32: false
 tokenizer_type: AutoTokenizer
@@ -89,10 +90,10 @@ use_accelerate: true
 val_set_size: 0.05
 wandb_entity: null
 wandb_mode: online
-wandb_name: c45db162-924b-42af-bc3a-aa67e6972820
+wandb_name: 12e79e69-9871-48aa-a364-52587c3c39bc
 wandb_project: Gradients-On-Demand
 wandb_run: your_name
-wandb_runid: c45db162-924b-42af-bc3a-aa67e6972820
+wandb_runid: 12e79e69-9871-48aa-a364-52587c3c39bc
 warmup_steps: 10
 weight_decay: 0.0
 xformers_attention: null
@@ -101,11 +102,11 @@ xformers_attention: null
 
 </details><br>
 
-# 85a3b62a-09c1-44d6-af05-0f30aee76da5
+# df5493b9-6228-4cae-a15d-b3a2fd8fb6e3
 
-This model is a fine-tuned version of [TitanML/tiny-mixtral](https://huggingface.co/TitanML/tiny-mixtral) on the Aivesa/dataset_c45db162-924b-42af-bc3a-aa67e6972820 dataset.
+This model is a fine-tuned version of [EleutherAI/pythia-70m-deduped](https://huggingface.co/EleutherAI/pythia-70m-deduped) on the Aivesa/dataset_12e79e69-9871-48aa-a364-52587c3c39bc dataset.
 It achieves the following results on the evaluation set:
-- Loss: 10.5178
+- Loss: 11.4571
 
 ## Model description
 
@@ -139,9 +140,9 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch  | Step | Validation Loss |
 |:-------------:|:------:|:----:|:---------------:|
-| 10.6162       | 0.0060 | 3    | 10.5356         |
-| 10.5933       | 0.0120 | 6    | 10.5323         |
-| 10.5445       | 0.0181 | 9    | 10.5178         |
+| 31.9198       | 0.0001 | 3    | 11.5754         |
+| 41.8347       | 0.0002 | 6    | 11.5426         |
+| 63.7696       | 0.0003 | 9    | 11.4571         |
 
 
 ### Framework versions