End of training

README.md

@@ -64,7 +64,7 @@ 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/b490c52030b0c7be_train_data.json
 model_type: AutoModelForCausalLM
@@ -91,7 +91,7 @@ wandb_name: 96d5bd4b-8c0a-4d7e-ba8b-9c0e2bd6dda6
 wandb_project: Gradients-On-Demand
 wandb_run: your_name
 wandb_runid: 96d5bd4b-8c0a-4d7e-ba8b-9c0e2bd6dda6
-warmup_steps: 1
+warmup_steps: 10
 weight_decay: 0.0
 xformers_attention: null
 
@@ -102,6 +102,8 @@ xformers_attention: null
 # 98e9fcdb-5dc3-4c9e-87cf-6d4c3ba3f193
 
 This model is a fine-tuned version of [EleutherAI/pythia-14m](https://huggingface.co/EleutherAI/pythia-14m) on the None dataset.
+It achieves the following results on the evaluation set:
+- Loss: 6.4142
 
 ## 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        | 0.0017 | 1    | 6.6529          |
+| 26.4966       | 0.0174 | 10   | 6.6043          |
+| 25.9504       | 0.0348 | 20   | 6.5139          |
+| 26.0941       | 0.0521 | 30   | 6.4433          |
+| 25.3607       | 0.0695 | 40   | 6.4201          |
+| 25.7918       | 0.0869 | 50   | 6.4142          |
 
 
 ### Framework versions

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f55b9ffb5ce1955c6dedb5253cec1cb35159d2f5fecd430488a789f4a5e1739a
+oid sha256:6ee83a35c030620bf7321d2725cc15580469a04b44aca30cfd5d522db77dd850
 size 410814