INDEX

Explanations

relevant phrases related to voting and legislative actions

oai_token-act-pair · gpt-3.5-turbo

mentions of voting or legislative actions

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

OGR

-0.79

ngth

-0.78

à¨

-0.73

vantage

-0.71

ACT

-0.71

uction

-0.70

oof

-0.69

ographer

-0.67

ographers

-0.66

IUM

-0.66

POSITIVE LOGITS

 whether

1.35

 behalf

1.27

 repealing

1.02

 Wednesday

0.96

 extending

0.95

 Tuesday

0.93

 Thursday

0.92

 authorizing

0.92

whether

0.91

 Monday

0.89

Activations Density 0.127%

relevant phrases related to voting and legislative actions

mentions of voting or legislative actions

No Comments

No Known Activations