INDEX

Explanations

German words and phrases

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.9.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

orsi

-0.80

Snake

-0.75

iles

-0.74

cemic

-0.74

eries

-0.72

earing

-0.71

akening

-0.68

aves

-0.68

Mex

-0.68

ilers

-0.68

POSITIVE LOGITS

 Zeit

1.00

 Bundes

0.95

 Volks

0.89

 Fors

0.88

 Ãľ

0.88

 alle

0.88

 polit

0.85

 Mens

0.84

mit

0.84

 vers

0.82

Activations Density 0.081%

German words and phrases

No Comments

No Known Activations