End of training

README.md

@@ -66,7 +66,7 @@ lora_model_dir: null
 lora_r: 8
 lora_target_linear: true
 lr_scheduler: cosine
-max_steps: 1
+max_steps: 50
 micro_batch_size: 2
 mlflow_experiment_name: /tmp/cc4db227b52147fc_train_data.json
 model_type: AutoModelForCausalLM
@@ -93,7 +93,7 @@ wandb_name: 4e3c4142-60c4-4a35-9720-e4d5fe7b97e0
 wandb_project: Gradients-On-Demand
 wandb_run: your_name
 wandb_runid: 4e3c4142-60c4-4a35-9720-e4d5fe7b97e0
-warmup_steps: 1
+warmup_steps: 10
 weight_decay: 0.0
 xformers_attention: null
 
@@ -104,6 +104,8 @@ xformers_attention: null
 # 2f7a7428-1081-4206-855c-4dc6417b4cbe
 
 This model is a fine-tuned version of [tiiuae/falcon-rw-1b](https://huggingface.co/tiiuae/falcon-rw-1b) on the None dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.3555
 
 ## Model description
 
@@ -130,14 +132,18 @@ The following hyperparameters were used during training:
 - 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: 2
-- training_steps: 1
+- lr_scheduler_warmup_steps: 10
+- training_steps: 22
 
 ### Training results
 
 | Training Loss | Epoch  | Step | Validation Loss |
 |:-------------:|:------:|:----:|:---------------:|
 | No log        | 0.0465 | 1    | 3.6673          |
+| 12.3052       | 0.2326 | 5    | 3.5459          |
+| 10.7092       | 0.4651 | 10   | 2.0259          |
+| 4.5191        | 0.6977 | 15   | 0.4031          |
+| 1.6795        | 0.9302 | 20   | 0.3555          |
 
 
 ### Framework versions

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5867fec5b9961b7ab16e8896fb8bf104704df95a0af963bdcfc3a231bb736a51
+oid sha256:6e22e9f9646df2b59767d46db4ba0b7acfc8d874ff2cdd61056274c6db8cd64c
 size 25236362