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/0898c0322684bbb1_train_data.json
 model_type: AutoModelForCausalLM
@@ -90,7 +90,7 @@ wandb_name: f39bf177-92c4-492f-9d24-29777d6903ee
 wandb_project: Gradients-On-Demand
 wandb_run: your_name
 wandb_runid: f39bf177-92c4-492f-9d24-29777d6903ee
-warmup_steps: 1
+warmup_steps: 10
 weight_decay: 0.0
 xformers_attention: null
 
@@ -101,6 +101,8 @@ xformers_attention: null
 # be092a14-1ae3-305c-153f-b9a314498942
 
 This model is a fine-tuned version of [katuni4ka/tiny-random-qwen1.5-moe](https://huggingface.co/katuni4ka/tiny-random-qwen1.5-moe) on the None dataset.
+It achieves the following results on the evaluation set:
+- Loss: 11.9224
 
 ## Model description
 
@@ -127,14 +129,19 @@ 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        | 0.0010 | 1    | 11.9379         |
+| 11.9368       | 0.0100 | 10   | 11.9359         |
+| 11.9319       | 0.0200 | 20   | 11.9308         |
+| 11.9255       | 0.0301 | 30   | 11.9259         |
+| 11.9256       | 0.0401 | 40   | 11.9230         |
+| 11.9223       | 0.0501 | 50   | 11.9224         |
 
 
 ### Framework versions

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:904155616cbbb8c3c3b2b01f45fec06a407210957fd7d1562305c5f201ecff52
+oid sha256:dfc5b462332841b074f5800e95758cd309b90c2ecb852005bab833fd78886e36
 size 388266