Mistral_ET_1

jlpan/Mistral_ET_1

Mistral_ET_1 is a machine learning model from jlpan. Use it for the machine learning task on the model card, and read the license before you ship it in a product. The card lists the license as apache-2.0.

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

Downloads · 30 days

Access

Public

Updated Oct 9, 2023

Repo size

274 MB

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.bin27.4 MB · 100%

At a glance

License: apache-2.0
Access: Public
Created: Oct 9, 2023
Updated: Oct 9, 2023
SHA: 46566871

Base models

mistralai/Mistral-7B-Instruct-v0.1

Model card

From the Hugging Face model README

Mistral_ET_1

This model is a fine-tuned version of mistralai/Mistral-7B-Instruct-v0.1 on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 0.7456

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: 5e-05
train_batch_size: 1
eval_batch_size: 1
seed: 42
gradient_accumulation_steps: 4
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: 50
training_steps: 1000

Training results

Training Loss	Epoch	Step	Validation Loss
2.5079	0.41	100	1.6764
1.3602	0.82	200	1.1685
1.0132	1.23	300	0.9737
0.8592	1.65	400	0.8837
0.7742	2.06	500	0.8322
0.6621	2.47	600	0.7871
0.6402	2.88	700	0.7554
0.5713	3.29	800	0.7530
0.5468	3.7	900	0.7472
0.5542	4.12	1000	0.7456

Framework versions

Transformers 4.34.0.dev0
Pytorch 2.0.1+cu117
Datasets 2.14.5
Tokenizers 0.14.0

License: apache-2.0
Created: Oct 9, 2023
Updated: Oct 9, 2023