Downloads · 30 days
4
27% of all-time downloads
mh_one_api
blur0b0t/mh_one_api
mh_one_api is a question answering model from blur0b0t. Use it when the input is a question plus a passage. It is set up for peft.
Flan-t5-xl finetuned model on dataset provided by intel|machinehack for QnA.
All-time downloads
15
Public
Repo size
57.8 MB
Likes
1
Public
Repo makeup
Click a slice to open those files.
.pt38 MB · 60%
At a glance
- Task
- Question Answering
- Library
- peft
- Access
- Public
- Created
- Oct 15, 2023
- Updated
- Oct 25, 2023
- SHA
- 2664735d
Try a prompt
Model card
From the Hugging Face model README
Model description
Flan-t5-xl finetuned model on dataset provided by intel|machinehack for QnA.
Intended uses & limitations
More information needed
Training hyperparameters
The following hyperparameters were used during training:
--per_device_train_batch_size 2
--per_device_eval_batch_size 2
--gradient_accumulation_steps 1
--do_train
--learning_rate 1.0e-5
--warmup_ratio 0.03
--weight_decay 0.0
--num_train_epochs 1 \
Training results
{ "epoch": 1.08, "train_loss": 0.0, "train_runtime": 0.017, "train_samples": 66611, "train_samples_per_second": 3929048.542, "train_steps_per_second": 1964553.764 }
Framework versions
- PEFT 0.5.0
- Task
- Question Answering
- Library
- peft
- Created
- Oct 15, 2023
- Updated
- Oct 25, 2023