imdb-gpt2-a2b1l0.0001

asun17904/imdb-gpt2-a2b1l0.0001

imdb-gpt2-a2b1l0.0001 is a machine learning model from asun17904. 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 pytorch. The card lists the license as mit.

Trainer Hyperparameters: - lr = 5e-05 - perdevicebatchsize = 8 - gradientaccumulationsteps = 2 - weightdecay = 1e-09 - seed = 42

Downloads · 30 days

15% of all-time downloads

All-time downloads

Public

Repo size

5 GB

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.bin498 MB · 100%

At a glance

Library: pytorch
License: mit
Model type: gpt2
Access: Public
Created: Jan 28, 2024
Updated: Jan 29, 2024
SHA: d09466ec

Model card

From the Hugging Face model README

Knowledge Continuity Regularized Network

Trainer Hyperparameters:

lr = 5e-05
per_device_batch_size = 8
gradient_accumulation_steps = 2
weight_decay = 1e-09
seed = 42

Regularization Hyperparameters

numerical stability denominator constant = 0.01
lambda = 0.0001
alpha = 2.0
beta = 1.0

Extended Logs:

eval_loss	eval_accuracy	epoch
12.577	0.916	1.0
14.443	0.857	2.0
12.267	0.928	3.0
12.056	0.934	4.0
12.092	0.934	5.0
12.172	0.932	6.0
12.208	0.930	7.0
12.934	0.908	8.0
11.949	0.939	9.0
12.090	0.935	10.0
12.315	0.927	11.0
12.018	0.937	12.0
12.001	0.938	13.0
12.128	0.934	14.0
12.064	0.935	15.0
11.960	0.938	16.0
12.116	0.934	17.0
12.313	0.928	18.0
11.999	0.938	19.0
11.996	0.938	20.0
12.019	0.937	21.0
12.002	0.938	22.0
12.180	0.932	23.0
11.963	0.939	24.0
12.045	0.936	25.0
11.942	0.940	26.0
12.153	0.933	27.0
11.840	0.943	28.0
11.905	0.941	29.0
11.899	0.941	30.0
11.854	0.942	31.0
11.972	0.939	32.0
11.922	0.940	33.0
11.922	0.940	34.0
11.894	0.941	35.0
11.878	0.942	36.0
11.853	0.942	37.0
11.914	0.940	38.0
11.867	0.942	39.0
11.862	0.942	40.0
11.855	0.942	41.0
11.900	0.941	42.0
11.828	0.943	43.0
11.853	0.942	44.0
11.847	0.943	45.0
11.867	0.942	46.0
11.868	0.942	47.0
11.838	0.943	48.0
11.832	0.943	49.0

Library: pytorch
Type: gpt2
License: mit
Languages: en
Created: Jan 28, 2024
Updated: Jan 29, 2024