End of training

README.md

@@ -1,14 +1,14 @@
 ---
 library_name: peft
-license: apache-2.0
-base_model: TinyLlama/TinyLlama_v1.1
+license: gemma
+base_model: unsloth/gemma-2-2b-it
 tags:
 - axolotl
 - generated_from_trainer
 datasets:
-- Paladiso/dataset_4a99212a-ed04-4c6f-b7e3-a8796dfe5605
+- Paladiso/dataset_e0ed89fb-80e3-4383-8dd1-e5d47cb56894
 model-index:
-- name: b77db938-be5a-42dc-a716-d6f0670735e9
+- name: a31e9467-1b84-42f8-a0e6-2c16bbde2ff7
   results: []
 ---
 
@@ -21,16 +21,15 @@ should probably proofread and complete it, then remove this comment. -->
 axolotl version: `0.6.0`
 ```yaml
 adapter: lora
-base_model: TinyLlama/TinyLlama_v1.1
+base_model: unsloth/gemma-2-2b-it
 bf16: auto
 chat_template: llama3
 dataset_prepared_path: /workspace/axolotl/data/prepared
 datasets:
 - ds_type: json
   format: custom
-  path: Paladiso/dataset_4a99212a-ed04-4c6f-b7e3-a8796dfe5605
+  path: Paladiso/dataset_e0ed89fb-80e3-4383-8dd1-e5d47cb56894
   type:
-    field_input: ''
     field_instruction: instruction
     field_output: output
     system_format: '{system}'
@@ -48,7 +47,7 @@ fsdp_config: null
 gradient_accumulation_steps: 4
 gradient_checkpointing: false
 group_by_length: false
-hub_model_id: Paladiso/b77db938-be5a-42dc-a716-d6f0670735e9
+hub_model_id: Paladiso/a31e9467-1b84-42f8-a0e6-2c16bbde2ff7
 hub_private_repo: true
 hub_repo: null
 hub_strategy: checkpoint
@@ -79,8 +78,6 @@ 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
@@ -90,10 +87,10 @@ use_accelerate: true
 val_set_size: 0.05
 wandb_entity: null
 wandb_mode: online
-wandb_name: 4a99212a-ed04-4c6f-b7e3-a8796dfe5605
+wandb_name: e0ed89fb-80e3-4383-8dd1-e5d47cb56894
 wandb_project: Gradients-On-Demand
 wandb_run: your_name
-wandb_runid: 4a99212a-ed04-4c6f-b7e3-a8796dfe5605
+wandb_runid: e0ed89fb-80e3-4383-8dd1-e5d47cb56894
 warmup_steps: 10
 weight_decay: 0.0
 xformers_attention: null
@@ -102,11 +99,11 @@ xformers_attention: null
 
 </details><br>
 
-# b77db938-be5a-42dc-a716-d6f0670735e9
+# a31e9467-1b84-42f8-a0e6-2c16bbde2ff7
 
-This model is a fine-tuned version of [TinyLlama/TinyLlama_v1.1](https://huggingface.co/TinyLlama/TinyLlama_v1.1) on the Paladiso/dataset_4a99212a-ed04-4c6f-b7e3-a8796dfe5605 dataset.
+This model is a fine-tuned version of [unsloth/gemma-2-2b-it](https://huggingface.co/unsloth/gemma-2-2b-it) on the Paladiso/dataset_e0ed89fb-80e3-4383-8dd1-e5d47cb56894 dataset.
 It achieves the following results on the evaluation set:
-- Loss: 3.1746
+- Loss: 1.3201
 
 ## Model description
 
@@ -140,9 +137,9 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch  | Step | Validation Loss |
 |:-------------:|:------:|:----:|:---------------:|
-| 2.688         | 0.0005 | 3    | 3.3686          |
-| 2.7628        | 0.0010 | 6    | 3.2964          |
-| 4.2762        | 0.0015 | 9    | 3.1746          |
+| 1.3971        | 0.0009 | 3    | 1.5252          |
+| 1.4187        | 0.0018 | 6    | 1.4485          |
+| 1.34          | 0.0026 | 9    | 1.3201          |
 
 
 ### Framework versions