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/59dfd08cf3fa50ec_train_data.json
 model_type: AutoModelForCausalLM
@@ -91,7 +91,7 @@ wandb_name: 9d1868e2-61ae-4041-af04-bf2ce6704f58
 wandb_project: Gradients-On-Demand
 wandb_run: your_name
 wandb_runid: 9d1868e2-61ae-4041-af04-bf2ce6704f58
-warmup_steps: 1
+warmup_steps: 10
 weight_decay: 0.0
 xformers_attention: null
 
@@ -102,6 +102,8 @@ xformers_attention: null
 # a64d677d-c2ba-4d52-a26c-33c32b9b9eea
 
 This model is a fine-tuned version of [unsloth/mistral-7b-instruct-v0.2](https://huggingface.co/unsloth/mistral-7b-instruct-v0.2) on the None dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.2857
 
 ## 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        | 0.0001 | 1    | 6.2291          |
+| 17.8844       | 0.0006 | 10   | 2.7177          |
+| 3.6848        | 0.0012 | 20   | 0.6884          |
+| 1.6636        | 0.0018 | 30   | 0.3703          |
+| 1.4148        | 0.0024 | 40   | 0.2983          |
+| 1.1832        | 0.0030 | 50   | 0.2857          |
 
 
 ### Framework versions

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:78acc7bacc61855f6507b603afaa8f664868cba8939d75cb5ef692e6234026de
+oid sha256:d51aa494d13c0a068ed191c0bb27e2653745449ac810bd4bd4dad3fa42790845
 size 84047370