INDEX

Explanations

phrases indicating a certain level or quality of something

oai_token-act-pair · gpt-3.5-turbo

references to a concept of "certain" or unspecified qualities or conditions

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

enders

-0.82

TeX

-0.78

POL

-0.76

IVERS

-0.76

 Cosponsors

-0.73

END

-0.72

ults

-0.71

inson

-0.71

MSN

-0.70

âĵĺ

-0.70

POSITIVE LOGITS

 amount

1.23

ties

1.12

 subset

1.06

 percentage

0.98

 degree

0.86

 extent

0.85

 threshold

0.83

 kind

0.80

 number

0.80

 piece

0.77

Activations Density 0.020%

phrases indicating a certain level or quality of something

references to a concept of "certain" or unspecified qualities or conditions

No Comments

No Known Activations