End of training

README.md

@@ -65,7 +65,7 @@ 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/22a7382f476a3340_train_data.json
 model_type: AutoModelForCausalLM
@@ -92,7 +92,7 @@ wandb_name: a1cb988f-7957-4a81-8cc4-c6f235d31d98
 wandb_project: Gradients-On-Demand
 wandb_run: your_name
 wandb_runid: a1cb988f-7957-4a81-8cc4-c6f235d31d98
-warmup_steps: 1
+warmup_steps: 10
 weight_decay: 0.0
 xformers_attention: null
 
@@ -103,6 +103,8 @@ xformers_attention: null
 # eb535eab-9459-4206-ba3b-6a7003a6d1fd
 
 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: 1.4932
 
 ## Model description
 
@@ -129,14 +131,19 @@ 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: 50
 
 ### Training results
 
 | Training Loss | Epoch  | Step | Validation Loss |
 |:-------------:|:------:|:----:|:---------------:|
 | No log        | 0.0002 | 1    | 1.7016          |
+| 6.6052        | 0.0024 | 10   | 1.6462          |
+| 6.5683        | 0.0048 | 20   | 1.5567          |
+| 6.1798        | 0.0073 | 30   | 1.5153          |
+| 6.1513        | 0.0097 | 40   | 1.4963          |
+| 5.9249        | 0.0121 | 50   | 1.4932          |
 
 
 ### Framework versions

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:65895b1d3354721c16d65f5f021ae82428de2da1e09b52aeee439e43e3395c1c
+oid sha256:c64254ad74c02c1aa22c7cf719a36b334511a7d92acf3fabc598f7626c3e894d
 size 25236362