hotpotqa-unsloth-mistral-7b-4bit-1024

12yuens2/hotpotqa-unsloth-mistral-7b-4bit-1024

hotpotqa-unsloth-mistral-7b-4bit-1024 is a machine learning model from 12yuens2. 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 apache-2.0.

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

Downloads · 30 days

5% of all-time downloads

All-time downloads

Public

Repo size

84.4 MB

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.safetensors83.9 MB · 97%

At a glance

Library: peft
License: apache-2.0
Access: Public
Created: Apr 26, 2024
Updated: Apr 26, 2024
SHA: e8af1960

Base models

unsloth/mistral-7b-instruct-v0.2-bnb-4bit

Model card

From the Hugging Face model README

mistral-7b-instruct-v0.2-bnb-4bit1024

This model is a fine-tuned version of unsloth/mistral-7b-instruct-v0.2-bnb-4bit on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 0.6953

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: 8
eval_batch_size: 8
seed: 42
gradient_accumulation_steps: 8
total_train_batch_size: 64
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_ratio: 0.05
num_epochs: 1

Training results

Training Loss	Epoch	Step	Validation Loss
1.8431	0.02	25	1.4131
0.8021	0.04	50	0.7911
0.7972	0.05	75	0.7886
0.7886	0.07	100	0.7780
0.7762	0.09	125	0.7546
0.7338	0.11	150	0.7332
0.707	0.12	175	0.7399
0.7252	0.14	200	0.7303
0.7513	0.16	225	0.7384
0.7275	0.18	250	0.7380
0.7283	0.19	275	0.7285
0.7132	0.21	300	0.7452
0.7273	0.23	325	0.7370
0.7353	0.25	350	0.7388
0.7457	0.27	375	0.7292
0.7404	0.28	400	0.7315
0.7312	0.3	425	0.7341
0.7285	0.32	450	0.7277
0.7331	0.34	475	0.7318
0.7179	0.35	500	0.7401
0.7432	0.37	525	0.7399
0.7305	0.39	550	0.7463
0.723	0.41	575	0.7448
0.7303	0.42	600	0.7339
0.7213	0.44	625	0.7320
0.7236	0.46	650	0.7378
0.7263	0.48	675	0.7451
0.7462	0.5	700	0.7238
0.7287	0.51	725	0.7274
0.7364	0.53	750	0.7369
0.7276	0.55	775	0.7282
0.7268	0.57	800	0.7431
0.7382	0.58	825	0.7376
0.7185	0.6	850	0.7402
0.7153	0.62	875	0.7362
0.7314	0.64	900	0.7395
0.7465	0.65	925	0.7378
0.7228	0.67	950	0.7333
0.7336	0.69	975	0.7337
0.72	0.71	1000	0.7313
0.7258	0.73	1025	0.7379
0.7312	0.74	1050	0.7342
0.7268	0.76	1075	0.7350
0.7137	0.78	1100	0.7401
0.7277	0.8	1125	0.7277
0.7314	0.81	1150	0.7388
0.7106	0.83	1175	0.7371
0.7226	0.85	1200	0.7326
0.7262	0.87	1225	0.7328
0.7356	0.88	1250	0.7408
0.7245	0.9	1275	0.7365
0.7221	0.92	1300	0.7404
0.7194	0.94	1325	0.7418
0.7209	0.96	1350	0.7380
0.7205	0.97	1375	0.7279
0.6788	0.99	1400	0.6953

Framework versions

PEFT 0.9.0
Transformers 4.38.2
Pytorch 2.2.1
Datasets 2.18.0
Tokenizers 0.15.1

Library: peft
License: apache-2.0
Created: Apr 26, 2024
Updated: Apr 26, 2024