Summary4500_M2_1000steps_1e6rate_SFT
This model is a fine-tuned version of mistralai/Mistral-7B-Instruct-v0.2 on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.4077
Model description
More information needed
Intended uses & limitations
More information needed
Training and evaluation data
More information needed
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 1e-06
- train_batch_size: 2
- eval_batch_size: 1
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 4
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 100
- training_steps: 1000
Training results
Training Loss | Epoch | Step | Validation Loss |
---|---|---|---|
0.4267 | 0.0447 | 50 | 0.4486 |
0.4244 | 0.0895 | 100 | 0.4530 |
0.4172 | 0.1342 | 150 | 0.4353 |
0.4055 | 0.1790 | 200 | 0.4316 |
0.4066 | 0.2237 | 250 | 0.4294 |
0.3974 | 0.2685 | 300 | 0.4279 |
0.4245 | 0.3132 | 350 | 0.4248 |
0.4032 | 0.3579 | 400 | 0.4211 |
0.4038 | 0.4027 | 450 | 0.4179 |
0.388 | 0.4474 | 500 | 0.4155 |
0.3814 | 0.4922 | 550 | 0.4132 |
0.3839 | 0.5369 | 600 | 0.4112 |
0.379 | 0.5817 | 650 | 0.4101 |
0.3887 | 0.6264 | 700 | 0.4091 |
0.4026 | 0.6711 | 750 | 0.4084 |
0.3903 | 0.7159 | 800 | 0.4079 |
0.3754 | 0.7606 | 850 | 0.4078 |
0.39 | 0.8054 | 900 | 0.4077 |
0.3917 | 0.8501 | 950 | 0.4077 |
0.3771 | 0.8949 | 1000 | 0.4077 |
Framework versions
- Transformers 4.42.3
- Pytorch 2.0.0+cu117
- Datasets 2.20.0
- Tokenizers 0.19.1
- Downloads last month
- 2
Model is too large to load in Inference API (serverless). To try the model, launch it on Inference Endpoints (dedicated)
instead.