INDEX

Explanations

words related to attitudes or beliefs

oai_token-act-pair · gpt-3.5-turbo

references to attitudes and opinions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

enegger

-0.87

icles

-0.83

oval

-0.76

dry

-0.74

detail

-0.73

eri

-0.72

aban

-0.71

orah

-0.71

oppable

-0.69

lan

-0.69

POSITIVE LOGITS

 toward

1.62

 towards

1.57

 Towards

1.29

 attitude

1.16

Tow

1.07

 attitudes

0.89

 trait

0.81

vis

0.79

 shift

0.78

 stance

0.78

Activations Density 0.053%

words related to attitudes or beliefs

references to attitudes and opinions

No Comments

No Known Activations