Mistral_Sparse_refined_web_50p_2024-03-29

thrunlab/Mistral_Sparse_refined_web_50p_2024-03-29

Mistral_Sparse_refined_web_50p_2024-03-29 is a text generation model from thrunlab. Use it when you need the model to write or continue text. It is set up for transformers. The card lists the license as apache-2.0.

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

Downloads · 30 days

9% of all-time downloads

All-time downloads

356

Public

Parameters

7.2B

121 GB on disk

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.safetensors14.8 GB · 100%

At a glance

Task: Text Generation
Library: transformers
License: apache-2.0
Model type: sparse_mistral
Access: Public
Created: Mar 29, 2024
Updated: Mar 29, 2024
SHA: 615dfcbb

Try a prompt

Base models

mistralai/Mistral-7B-v0.1

Model card

From the Hugging Face model README

Mistral_Sparse_refined_web_50p_2024-03-29

This model is a fine-tuned version of mistralai/Mistral-7B-v0.1 on the None dataset. It achieves the following results on the evaluation set:

Loss: 2.3556

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-05
train_batch_size: 1
eval_batch_size: 1
seed: 0
distributed_type: multi-GPU
num_devices: 4
gradient_accumulation_steps: 4
total_train_batch_size: 16
total_eval_batch_size: 4
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
training_steps: 1600

Training results

Training Loss	Epoch	Step	Validation Loss
2.4102	0.0	25	2.6367
2.4844	0.01	50	2.5737
2.3601	0.01	75	2.5236
2.4524	0.02	100	2.4934
2.3586	0.02	125	2.4753
2.3004	0.03	150	2.4581
2.2308	0.03	175	2.4502
2.3163	0.04	200	2.4406
2.2026	0.04	225	2.4344
2.226	0.05	250	2.4264
2.3057	0.05	275	2.4255
2.288	0.06	300	2.4206
2.2401	0.06	325	2.4145
2.2386	0.07	350	2.4100
2.2824	0.07	375	2.4115
2.2264	0.08	400	2.4107
2.1135	0.08	425	2.4099
2.4512	0.09	450	2.4079
2.2329	0.09	475	2.4077
2.2044	0.09	500	2.4042
2.2461	0.1	525	2.4053
2.2781	0.1	550	2.4035
2.3529	0.11	575	2.4003
2.3395	0.11	600	2.4021
2.212	0.12	625	2.4006
2.344	0.12	650	2.4004
2.1826	0.13	675	2.4015
2.1783	0.13	700	2.3991
2.1863	0.14	725	2.3952
2.2943	0.14	750	2.3952
2.2353	0.15	775	2.3940
2.3216	0.15	800	2.3959
2.2007	0.16	825	2.3935
2.1674	0.16	850	2.3950
2.2554	0.17	875	2.3952
2.2562	0.17	900	2.3932
2.2707	0.17	925	2.3939
2.3025	0.18	950	2.3895
2.3242	0.18	975	2.3899
2.2643	0.19	1000	2.3903
2.2352	0.19	1025	2.3895
2.2249	0.2	1050	2.3883
2.1267	0.2	1075	2.3878
2.2937	0.21	1100	2.3880
2.2461	0.21	1125	2.3870
2.2747	0.22	1150	2.3891
2.1852	0.22	1175	2.3873
2.2889	0.23	1200	2.3866
2.1805	0.23	1225	2.3863
2.1848	0.24	1250	2.3885
2.2034	0.24	1275	2.3875
2.2227	0.25	1300	2.3875
2.3218	0.25	1325	2.3852
2.3232	0.26	1350	2.3842
2.2248	0.26	1375	2.3874
2.3093	0.26	1400	2.3821
2.3094	0.27	1425	2.3849
2.2672	0.27	1450	2.3820
2.2513	0.28	1475	2.3801
2.3726	0.28	1500	2.3799
2.1227	0.29	1525	2.3807
2.3133	0.29	1550	2.3807
2.3258	0.3	1575	2.3802
2.23	0.3	1600	2.3829

Framework versions

Transformers 4.36.2
Pytorch 2.1.2+cu121
Datasets 2.15.0
Tokenizers 0.15.0

Task: Text Generation
Library: transformers
Type: sparse_mistral
License: apache-2.0
Created: Mar 29, 2024
Updated: Mar 29, 2024