llama_7b_qlora_pds-eval

DevaMalla/llama_7b_qlora_pds-eval

llama_7b_qlora_pds-eval is a machine learning model from DevaMalla. Use it for the machine learning task on the model card, and read the license before you ship it in a product.

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

Downloads · 30 days

Access

Public

Updated Sep 22, 2023

Repo size

134 MB

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.bin134 MB · 100%

At a glance

Access: Public
Created: Sep 21, 2023
Updated: Sep 22, 2023
SHA: d20419e1

Base models

DevaMalla/llama7b

Model card

From the Hugging Face model README

llama_7b_qlora_pds-eval

This model is a fine-tuned version of DevaMalla/llama7b on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 0.9304

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: 0.0002
train_batch_size: 10
eval_batch_size: 2
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 1

Training results

Training Loss	Epoch	Step	Validation Loss
1.2419	0.03	50	1.0315
0.9965	0.05	100	0.9918
0.9747	0.08	150	0.9787
0.9771	0.11	200	0.9705
0.9653	0.14	250	0.9645
0.9507	0.16	300	0.9599
0.9596	0.19	350	0.9563
0.9497	0.22	400	0.9542
0.9481	0.24	450	0.9517
0.9475	0.27	500	0.9495
0.9376	0.3	550	0.9476
0.943	0.33	600	0.9463
0.952	0.35	650	0.9447
0.9489	0.38	700	0.9433
0.9347	0.41	750	0.9421
0.9307	0.44	800	0.9414
0.9366	0.46	850	0.9406
0.9246	0.49	900	0.9393
0.9267	0.52	950	0.9383
0.9358	0.54	1000	0.9373
0.9405	0.57	1050	0.9365
0.9276	0.6	1100	0.9359
0.9403	0.63	1150	0.9353
0.9218	0.65	1200	0.9347
0.9406	0.68	1250	0.9343
0.9257	0.71	1300	0.9337
0.92	0.73	1350	0.9334
0.9356	0.76	1400	0.9328
0.9279	0.79	1450	0.9322
0.9214	0.82	1500	0.9320
0.9214	0.84	1550	0.9316
0.9258	0.87	1600	0.9312
0.9308	0.9	1650	0.9310
0.9251	0.93	1700	0.9307
0.9207	0.95	1750	0.9306
0.926	0.98	1800	0.9304

Framework versions

Transformers 4.32.0.dev0
Pytorch 2.0.0
Datasets 2.12.0
Tokenizers 0.13.3

Created: Sep 21, 2023
Updated: Sep 22, 2023