themis-instruct

gsoaresbaptista/themis-instruct

themis-instruct is a machine learning model from gsoaresbaptista. Use it for the machine learning task on the model card, and read the license before you ship it in a product. It is set up for peft. The card lists the license as llama2.

should probably proofread and complete it, then remove this comment. --

Downloads · 30 days

34% of all-time downloads

All-time downloads

Public

Repo size

1.7 GB

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.safetensors849 MB · 100%

At a glance

Library: peft
License: llama2
Access: Public
Created: Dec 19, 2023
Updated: Dec 19, 2023
SHA: 60e5736e

Base models

dominguesm/canarim-7b-instruct

Model card

From the Hugging Face model README

themis-instruct

This model is a fine-tuned version of dominguesm/canarim-7b-instruct on the None dataset. It achieves the following results on the evaluation set:

Loss: 0.8855

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: 2.5e-05
train_batch_size: 16
eval_batch_size: 8
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: cosine
lr_scheduler_warmup_steps: 1
num_epochs: 1

Training results

Training Loss	Epoch	Step	Validation Loss
2.0042	0.03	25	1.3872
1.2414	0.05	50	1.1247
1.0936	0.08	75	1.0210
1.0607	0.11	100	0.9881
1.0186	0.13	125	0.9693
0.9851	0.16	150	0.9552
1.0123	0.18	175	0.9469
0.9737	0.21	200	0.9398
0.9408	0.24	225	0.9371
0.9129	0.26	250	0.9300
0.9441	0.29	275	0.9254
0.9577	0.32	300	0.9198
0.9509	0.34	325	0.9162
0.9023	0.37	350	0.9135
0.8924	0.4	375	0.9109
0.9207	0.42	400	0.9086
0.9436	0.45	425	0.9058
0.8637	0.47	450	0.9047
0.9207	0.5	475	0.9033
0.9475	0.53	500	0.9002
0.9548	0.55	525	0.8981
0.8806	0.58	550	0.8969
0.9475	0.61	575	0.8949
0.8505	0.63	600	0.8932
0.8999	0.66	625	0.8926
0.9018	0.68	650	0.8906
0.9107	0.71	675	0.8901
0.8557	0.74	700	0.8888
0.8903	0.76	725	0.8881
0.8718	0.79	750	0.8875
0.9002	0.82	775	0.8870
0.9086	0.84	800	0.8867
0.8983	0.87	825	0.8863
0.9401	0.9	850	0.8861
0.9434	0.92	875	0.8857
0.8987	0.95	900	0.8855
0.9008	0.97	925	0.8855

Framework versions

PEFT 0.7.2.dev0
Transformers 4.37.0.dev0
Pytorch 2.1.0+cu121
Datasets 2.15.0
Tokenizers 0.15.0

Library: peft
License: llama2
Created: Dec 19, 2023
Updated: Dec 19, 2023