INDEX

Explanations

phrases related to beliefs or speculations

oai_token-act-pair · gpt-3.5-turbo

phrases indicating belief or speculation about the status or actions of individuals or entities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

 Perspective

-0.73

bats

-0.69

 Plain

-0.66

blooded

-0.65

case

-0.63

holder

-0.60

noticed

-0.60

unciation

-0.59

loads

-0.59

 horizont

-0.59

POSITIVE LOGITS

 indicate

0.95

be

0.93

 imply

0.89

 resemble

0.88

 accelerate

0.87

asted

0.86

 signify

0.84

 differentiate

0.83

 embody

0.83

 incorporate

0.82

Activations Density 0.103%

phrases related to beliefs or speculations

phrases indicating belief or speculation about the status or actions of individuals or entities

No Comments

No Known Activations