Downloads · 30 days
4
5% of all-time downloads
gLM2-1protonly-3epoch-finetune
ishanjmukherjee/gLM2-1protonly-3epoch-finetune
gLM2-1protonly-3epoch-finetune is a fill-mask model from ishanjmukherjee. Use it when you need the model to fill a missing word. It is set up for transformers.
should probably proofread and complete it, then remove this comment. --
All-time downloads
75
Public
Parameters
671M
153 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors2.7 GB · 100%
At a glance
- Task
- Fill-Mask
- Library
- transformers
- Model type
- gLM2
- Access
- Public
- Created
- May 8, 2025
- Updated
- May 9, 2025
- SHA
- 3817887f
Try a prompt
Base models
Model card
From the Hugging Face model README
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
gLM2-1protonly-3epoch-finetune
This model is a fine-tuned version of tattabio/gLM2_650M on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.9921
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: 0.0001
- train_batch_size: 2
- eval_batch_size: 2
- seed: 42
- distributed_type: multi-GPU
- num_devices: 4
- total_train_batch_size: 8
- total_eval_batch_size: 8
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- num_epochs: 3
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 1.2307 | 0.0529 | 500 | 1.2231 |
| 1.1913 | 0.1059 | 1000 | 1.1909 |
| 1.1672 | 0.1588 | 1500 | 1.1665 |
| 1.162 | 0.2118 | 2000 | 1.1493 |
| 1.1274 | 0.2647 | 2500 | 1.1367 |
| 1.1217 | 0.3177 | 3000 | 1.1252 |
| 1.1159 | 0.3706 | 3500 | 1.1151 |
| 1.108 | 0.4235 | 4000 | 1.1049 |
| 1.095 | 0.4765 | 4500 | 1.0972 |
| 1.0907 | 0.5294 | 5000 | 1.0898 |
| 1.0894 | 0.5824 | 5500 | 1.0849 |
| 1.0903 | 0.6353 | 6000 | 1.0794 |
| 1.076 | 0.6883 | 6500 | 1.0745 |
| 1.0719 | 0.7412 | 7000 | 1.0685 |
| 1.0765 | 0.7942 | 7500 | 1.0639 |
| 1.081 | 0.8471 | 8000 | 1.0602 |
| 1.0497 | 0.9000 | 8500 | 1.0567 |
| 1.0567 | 0.9530 | 9000 | 1.0524 |
| 1.0395 | 1.0059 | 9500 | 1.0508 |
| 1.0401 | 1.0589 | 10000 | 1.0482 |
| 1.0434 | 1.1118 | 10500 | 1.0449 |
| 1.0632 | 1.1648 | 11000 | 1.0433 |
| 1.0256 | 1.2177 | 11500 | 1.0386 |
| 1.0253 | 1.2706 | 12000 | 1.0370 |
| 1.0303 | 1.3236 | 12500 | 1.0339 |
| 1.028 | 1.3765 | 13000 | 1.0311 |
| 1.0076 | 1.4295 | 13500 | 1.0290 |
| 1.0303 | 1.4824 | 14000 | 1.0272 |
| 1.0277 | 1.5354 | 14500 | 1.0244 |
| 1.0268 | 1.5883 | 15000 | 1.0222 |
| 1.0074 | 1.6413 | 15500 | 1.0207 |
| 1.0211 | 1.6942 | 16000 | 1.0196 |
| 1.0178 | 1.7471 | 16500 | 1.0176 |
| 1.0068 | 1.8001 | 17000 | 1.0161 |
| 1.0169 | 1.8530 | 17500 | 1.0144 |
| 1.026 | 1.9060 | 18000 | 1.0126 |
| 0.997 | 1.9589 | 18500 | 1.0109 |
| 0.9979 | 2.0119 | 19000 | 1.0098 |
| 1.0105 | 2.0648 | 19500 | 1.0084 |
| 0.9903 | 2.1177 | 20000 | 1.0077 |
| 1.0188 | 2.1707 | 20500 | 1.0052 |
| 1.005 | 2.2236 | 21000 | 1.0040 |
| 0.9967 | 2.2766 | 21500 | 1.0036 |
| 0.9914 | 2.3295 | 22000 | 1.0014 |
| 0.9845 | 2.3825 | 22500 | 1.0010 |
| 0.9974 | 2.4354 | 23000 | 0.9999 |
| 0.9949 | 2.4884 | 23500 | 0.9987 |
| 0.9829 | 2.5413 | 24000 | 0.9971 |
| 0.9882 | 2.5942 | 24500 | 0.9968 |
| 0.9931 | 2.6472 | 25000 | 0.9955 |
| 0.9964 | 2.7001 | 25500 | 0.9944 |
| 0.9975 | 2.7531 | 26000 | 0.9939 |
| 0.9824 | 2.8060 | 26500 | 0.9938 |
| 0.9752 | 2.8590 | 27000 | 0.9928 |
| 0.9757 | 2.9119 | 27500 | 0.9931 |
| 0.9694 | 2.9648 | 28000 | 0.9921 |
Framework versions
- Transformers 4.51.3
- Pytorch 2.7.0+cu126
- Datasets 3.5.1
- Tokenizers 0.21.1
- Task
- Fill-Mask
- Library
- transformers
- Type
- gLM2
- Created
- May 8, 2025
- Updated
- May 9, 2025