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/cd99aeb9bb20ac18_train_data.json
 model_type: AutoModelForCausalLM
@@ -91,7 +91,7 @@ wandb_name: c73ca2e3-4ca3-4c15-ab47-be50260d2e94
 wandb_project: Gradients-On-Demand
 wandb_run: your_name
 wandb_runid: c73ca2e3-4ca3-4c15-ab47-be50260d2e94
-warmup_steps: 1
+warmup_steps: 10
 weight_decay: 0.0
 xformers_attention: null
 
@@ -102,6 +102,8 @@ xformers_attention: null
 # f9efd1fd-5abe-bd31-6327-2f1b99f5d624
 
 This model is a fine-tuned version of [fxmarty/really-tiny-falcon-testing](https://huggingface.co/fxmarty/really-tiny-falcon-testing) on the None dataset.
+It achieves the following results on the evaluation set:
+- Loss: 11.0741
 
 ## Model description
 
@@ -128,14 +130,19 @@ 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        | 0.0018 | 1    | 11.0828         |
+| 44.3285       | 0.0176 | 10   | 11.0812         |
+| 44.3119       | 0.0351 | 20   | 11.0778         |
+| 44.2984       | 0.0527 | 30   | 11.0757         |
+| 44.2978       | 0.0703 | 40   | 11.0743         |
+| 44.296        | 0.0878 | 50   | 11.0741         |
 
 
 ### Framework versions

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0aaa78e38a004f6634c4fac715316e12971dbb9f87ead6bc53ad91ca107749ab
+oid sha256:1541e20a99eb4e1407dd51cd0d2aaba784dc2d6f87cae47dbbf9a50c8256d1c4
 size 37022