INDEX

Explanations

the word "true" with a high degree of activation

oai_token-act-pair · gpt-3.5-turbo

references to authenticity or being genuine

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

served

-0.76

uled

-0.75

RAW

-0.75

Pages

-0.74

acco

-0.74

ocene

-0.73

ONES

-0.73

Corp

-0.73

ambo

-0.71

chains

-0.69

POSITIVE LOGITS

 believers

0.90

 believer

0.88

ignment

0.73

 positives

0.69

itability

0.68

polit

0.68

 patriot

0.67

 freshman

0.67

ance

0.66

ll

0.66

Activations Density 0.018%

the word "true" with a high degree of activation

references to authenticity or being genuine

No Comments

No Known Activations