INDEX

Explanations

terms related to problematic situations or issues

oai_token-act-pair · gpt-3.5-turbo

words related to problems or difficulties

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

oliberal

-0.69

entric

-0.67

urse

-0.66

VO

-0.64

ulture

-0.64

ivist

-0.60

icrobial

-0.58

ortium

-0.57

eele

-0.57

ocide

-0.57

POSITIVE LOGITS

hooting

1.50

makers

1.24

maker

1.09

some

0.99

nel

0.86

making

0.85

fully

0.83

engers

0.82

nels

0.81

pots

0.80

Activations Density 0.060%

terms related to problematic situations or issues

words related to problems or difficulties

No Comments

No Known Activations