End of training

README.md

@@ -59,7 +59,7 @@ lora_model_dir: null
 lora_r: 8
 lora_target_linear: true
 lr_scheduler: cosine
-max_steps: 1
+max_steps: 50
 micro_batch_size: 8
 mlflow_experiment_name: /tmp/471c7eff63ae8cba_train_data.json
 model_type: AutoModelForCausalLM
@@ -84,7 +84,7 @@ wandb_name: 580fd833-be10-49bf-819f-96cd012e6af8
 wandb_project: Gradients-On-Demand
 wandb_run: your_name
 wandb_runid: 580fd833-be10-49bf-819f-96cd012e6af8
-warmup_steps: 1
+warmup_steps: 2
 weight_decay: 0.0
 xformers_attention: null
 
@@ -95,6 +95,8 @@ xformers_attention: null
 # 580fd833-be10-49bf-819f-96cd012e6af8
 
 This model is a fine-tuned version of [unsloth/OpenHermes-2.5-Mistral-7B](https://huggingface.co/unsloth/OpenHermes-2.5-Mistral-7B) on the None dataset.
+It achieves the following results on the evaluation set:
+- Loss: 1.0420
 
 ## Model description
 
@@ -122,13 +124,17 @@ The following hyperparameters were used during training:
 - 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
+- training_steps: 9
 
 ### Training results
 
 | Training Loss | Epoch  | Step | Validation Loss |
 |:-------------:|:------:|:----:|:---------------:|
 | No log        | 0.1119 | 1    | 1.4554          |
+| No log        | 0.2238 | 2    | 1.3605          |
+| No log        | 0.4476 | 4    | 1.1157          |
+| No log        | 0.6713 | 6    | 1.0689          |
+| No log        | 0.8951 | 8    | 1.0420          |
 
 
 ### Framework versions

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e31b418b9ea321f180420c9f679574bcd7ce694746a456111b419314e5364cdb
+oid sha256:154fd04952ef306d62b7fd62187861a8b22a6f8daff70627c4c0c50d05f391d9
 size 84047370