End of training

README.md

@@ -1,14 +1,14 @@
 ---
 library_name: peft
-license: apache-2.0
-base_model: unsloth/Qwen2.5-1.5B
+license: llama2
+base_model: codellama/CodeLlama-7b-hf
 tags:
 - axolotl
 - generated_from_trainer
 datasets:
-- Paladiso/dataset_7fc40eed-6bc5-4668-b817-a908f0a659fe
+- Paladiso/dataset_1e5e9609-ad29-4797-9b4d-ea610be96ec8
 model-index:
-- name: e2729802-dde6-43c2-bb4a-4d0a4beac90e
+- name: ffa06f47-6de0-4c5b-84ef-c8437166eacd
   results: []
 ---
 
@@ -21,16 +21,17 @@ should probably proofread and complete it, then remove this comment. -->
 axolotl version: `0.6.0`
 ```yaml
 adapter: lora
-base_model: unsloth/Qwen2.5-1.5B
+base_model: codellama/CodeLlama-7b-hf
 bf16: auto
 chat_template: llama3
 dataset_prepared_path: /workspace/axolotl/data/prepared
 datasets:
 - ds_type: json
   format: custom
-  path: Paladiso/dataset_7fc40eed-6bc5-4668-b817-a908f0a659fe
+  path: Paladiso/dataset_1e5e9609-ad29-4797-9b4d-ea610be96ec8
   type:
-    field_instruction: abstract
+    field_input: ''
+    field_instruction: content
     field_output: title
     system_format: '{system}'
     system_prompt: ''
@@ -47,7 +48,7 @@ fsdp_config: null
 gradient_accumulation_steps: 4
 gradient_checkpointing: false
 group_by_length: false
-hub_model_id: Paladiso/e2729802-dde6-43c2-bb4a-4d0a4beac90e
+hub_model_id: Paladiso/ffa06f47-6de0-4c5b-84ef-c8437166eacd
 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: </s>
 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: 7fc40eed-6bc5-4668-b817-a908f0a659fe
+wandb_name: 1e5e9609-ad29-4797-9b4d-ea610be96ec8
 wandb_project: Gradients-On-Demand
 wandb_run: your_name
-wandb_runid: 7fc40eed-6bc5-4668-b817-a908f0a659fe
+wandb_runid: 1e5e9609-ad29-4797-9b4d-ea610be96ec8
 warmup_steps: 10
 weight_decay: 0.0
 xformers_attention: null
@@ -99,11 +102,11 @@ xformers_attention: null
 
 </details><br>
 
-# e2729802-dde6-43c2-bb4a-4d0a4beac90e
+# ffa06f47-6de0-4c5b-84ef-c8437166eacd
 
-This model is a fine-tuned version of [unsloth/Qwen2.5-1.5B](https://huggingface.co/unsloth/Qwen2.5-1.5B) on the Paladiso/dataset_7fc40eed-6bc5-4668-b817-a908f0a659fe dataset.
+This model is a fine-tuned version of [codellama/CodeLlama-7b-hf](https://huggingface.co/codellama/CodeLlama-7b-hf) on the Paladiso/dataset_1e5e9609-ad29-4797-9b4d-ea610be96ec8 dataset.
 It achieves the following results on the evaluation set:
-- Loss: 2.3707
+- Loss: 2.2325
 
 ## Model description
 
@@ -137,9 +140,9 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch  | Step | Validation Loss |
 |:-------------:|:------:|:----:|:---------------:|
-| 2.327         | 0.0004 | 3    | 2.8300          |
-| 2.4           | 0.0007 | 6    | 2.7118          |
-| 2.8662        | 0.0011 | 9    | 2.3707          |
+| 2.2442        | 0.0117 | 3    | 2.4155          |
+| 1.9119        | 0.0234 | 6    | 2.3805          |
+| 2.6282        | 0.0352 | 9    | 2.2325          |
 
 
 ### Framework versions