INDEX

Explanations

names of specific car models

oai_token-act-pair · gpt-3.5-turbo

names and terms associated with high-end brands or luxury items

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

worm

-0.80

worker

-0.80

worms

-0.77

less

-0.75

lands

-0.74

workers

-0.72

box

-0.70

bread

-0.70

winner

-0.70

work

-0.69

POSITIVE LOGITS

iazep

0.97

aceae

0.92

éĹĺ

0.92

acca

0.91

cffffcc

0.89

borgh

0.87

umbn

0.86

itars

0.83

uyomi

0.80

dayName

0.78

Activations Density 0.050%

names of specific car models

names and terms associated with high-end brands or luxury items

No Comments

No Known Activations