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/504df86bb7a8a89f_train_data.json
 model_type: AutoModelForCausalLM
@@ -93,7 +93,7 @@ wandb_name: 4c4b82d3-69c9-4409-810d-7f97b4d097eb
 wandb_project: Gradients-On-Demand
 wandb_run: your_name
 wandb_runid: 4c4b82d3-69c9-4409-810d-7f97b4d097eb
-warmup_steps: 1
+warmup_steps: 10
 weight_decay: 0.0
 xformers_attention: null
 
@@ -104,6 +104,8 @@ xformers_attention: null
 # d64256d3-018f-ce7d-37e8-2a4fe8a0edbd
 
 This model is a fine-tuned version of [echarlaix/tiny-random-mistral](https://huggingface.co/echarlaix/tiny-random-mistral) on the None dataset.
+It achieves the following results on the evaluation set:
+- Loss: 10.3740
 
 ## Model description
 
@@ -130,14 +132,19 @@ 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        | 0.0001 | 1    | 10.3780         |
+| 41.5017       | 0.0007 | 10   | 10.3773         |
+| 41.516        | 0.0014 | 20   | 10.3759         |
+| 41.5069       | 0.0020 | 30   | 10.3747         |
+| 41.4965       | 0.0027 | 40   | 10.3741         |
+| 41.4887       | 0.0034 | 50   | 10.3740         |
 
 
 ### Framework versions

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c45aed109386423f116251d9e4dd5aad08c00731130a54fbbe84a82f864f30b5
+oid sha256:c784a44b2950b2f2215087164c37b3ee09829e3f2785f014e515b2a6602c9381
 size 37506