willtensora
/

00000000-0000-0000-0000-000000000000

PEFT

Safetensors

qwen2

axolotl

Generated from Trainer

Model card Files Files and versions Community

willtensora commited on 3 days ago

Commit

911b2a1

verified ·

1 Parent(s): aae77c7

End of training

Browse files

Files changed (2) hide show

README.md +32 -26
adapter_model.bin +2 -2

README.md CHANGED Viewed

@@ -19,7 +19,7 @@ axolotl version: `0.4.1`
 ```yaml
 adapter: lora
 base_model: peft-internal-testing/tiny-dummy-qwen2
-bf16: true
 chat_template: llama3
 dataset_prepared_path: null
 datasets:
@@ -36,34 +36,34 @@ datasets:
 debug: null
 deepspeed: null
 early_stopping_patience: null
-eval_max_new_tokens: 64
 eval_table_size: null
-evals_per_epoch: 0
 flash_attention: false
-fp16: false
 fsdp: null
 fsdp_config: null
-gradient_accumulation_steps: 1
 gradient_checkpointing: false
-group_by_length: true
 hub_model_id: willtensora/00000000-0000-0000-0000-000000000000
 hub_repo: null
 hub_strategy: checkpoint
 hub_token: null
-learning_rate: 0.001
 load_in_4bit: false
 load_in_8bit: false
 local_rank: null
 logging_steps: 1
-lora_alpha: 8
-lora_dropout: 0.1
 lora_fan_in_fan_out: null
 lora_model_dir: null
-lora_r: 4
 lora_target_linear: true
-lr_scheduler: linear
-max_steps: 1
-micro_batch_size: 4
 mlflow_experiment_name: argilla/databricks-dolly-15k-curated-en
 model_type: AutoModelForCausalLM
 num_epochs: 1
@@ -73,21 +73,21 @@ pad_to_sequence_len: true
 resume_from_checkpoint: null
 s2_attention: null
 sample_packing: false
-saves_per_epoch: 0
-sequence_len: 64
 strict: false
-tf32: true
 tokenizer_type: AutoTokenizer
 train_on_inputs: false
 trust_remote_code: true
-val_set_size: 0.001
 wandb_entity: null
-wandb_mode: disabled
 wandb_name: 00000000-0000-0000-0000-000000000000
 wandb_project: Gradients-On-Demand
 wandb_run: your_name
 wandb_runid: 00000000-0000-0000-0000-000000000000
-warmup_steps: 0
 weight_decay: 0.0
 xformers_attention: null
@@ -99,7 +99,7 @@ xformers_attention: null
 This model is a fine-tuned version of [peft-internal-testing/tiny-dummy-qwen2](https://huggingface.co/peft-internal-testing/tiny-dummy-qwen2) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 11.9339
 ## Model description
@@ -118,19 +118,25 @@ More information needed
 ### Training hyperparameters
 The following hyperparameters were used during training:
-- learning_rate: 0.001
-- train_batch_size: 4
-- eval_batch_size: 4
 - seed: 42
 - optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
-- lr_scheduler_type: linear
-- training_steps: 1
 ### Training results
 | Training Loss | Epoch  | Step | Validation Loss |
 |:-------------:|:------:|:----:|:---------------:|
-| 11.9308       | 0.0003 | 1    | 11.9339         |
 ### Framework versions

 ```yaml
 adapter: lora
 base_model: peft-internal-testing/tiny-dummy-qwen2
+bf16: auto
 chat_template: llama3
 dataset_prepared_path: null
 datasets:
 debug: null
 deepspeed: null
 early_stopping_patience: null
+eval_max_new_tokens: 128
 eval_table_size: null
+evals_per_epoch: 4
 flash_attention: false
+fp16: null
 fsdp: null
 fsdp_config: null
+gradient_accumulation_steps: 4
 gradient_checkpointing: false
+group_by_length: false
 hub_model_id: willtensora/00000000-0000-0000-0000-000000000000
 hub_repo: null
 hub_strategy: checkpoint
 hub_token: null
+learning_rate: 0.0002
 load_in_4bit: false
 load_in_8bit: false
 local_rank: null
 logging_steps: 1
+lora_alpha: 16
+lora_dropout: 0.05
 lora_fan_in_fan_out: null
 lora_model_dir: null
+lora_r: 8
 lora_target_linear: true
+lr_scheduler: cosine
+max_steps: 10
+micro_batch_size: 2
 mlflow_experiment_name: argilla/databricks-dolly-15k-curated-en
 model_type: AutoModelForCausalLM
 num_epochs: 1
 resume_from_checkpoint: null
 s2_attention: null
 sample_packing: false
+saves_per_epoch: 4
+sequence_len: 512
 strict: false
+tf32: false
 tokenizer_type: AutoTokenizer
 train_on_inputs: false
 trust_remote_code: true
+val_set_size: 0.05
 wandb_entity: null
+wandb_mode: online
 wandb_name: 00000000-0000-0000-0000-000000000000
 wandb_project: Gradients-On-Demand
 wandb_run: your_name
 wandb_runid: 00000000-0000-0000-0000-000000000000
+warmup_steps: 10
 weight_decay: 0.0
 xformers_attention: null
 This model is a fine-tuned version of [peft-internal-testing/tiny-dummy-qwen2](https://huggingface.co/peft-internal-testing/tiny-dummy-qwen2) on the None dataset.
 It achieves the following results on the evaluation set:
+- Loss: 11.9313
 ## Model description
 ### Training hyperparameters
 The following hyperparameters were used during training:
+- learning_rate: 0.0002
+- train_batch_size: 2
+- eval_batch_size: 2
 - seed: 42
+- gradient_accumulation_steps: 4
+- total_train_batch_size: 8
 - optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
+- lr_scheduler_type: cosine
+- lr_scheduler_warmup_steps: 10
+- training_steps: 10
 ### Training results
 | Training Loss | Epoch  | Step | Validation Loss |
 |:-------------:|:------:|:----:|:---------------:|
+| 11.9315       | 0.0006 | 1    | 11.9313         |
+| 11.9319       | 0.0017 | 3    | 11.9313         |
+| 11.926        | 0.0034 | 6    | 11.9313         |
+| 11.9287       | 0.0050 | 9    | 11.9313         |
 ### Framework versions

adapter_model.bin CHANGED Viewed

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:13de7e032039330bbf4f0c30a9567db314f04a579cbf86d7fe79735c5b784d05
-size 15746

 version https://git-lfs.github.com/spec/v1
+oid sha256:06971f09ee5207a9e18e33f4542c9d182faded39e00ca4445655a418cede3d5c
+size 21378