INDEX

Explanations

Twitter hashtags or promotional keywords

oai_token-act-pair · gpt-3.5-turbo

instances of the letter 'W' and associated patterns in text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

glers

-0.84

ĵĺ

-0.75

ï¸ı

-0.73

#$#$

-0.69

LOAD

-0.65

ongyang

-0.62

Reply

-0.59

GROUP

-0.58

ccording

-0.57

xtap

-0.57

POSITIVE LOGITS

hyde

0.87

ciating

0.82

igans

0.75

enium

0.69

enment

0.69

xia

0.69

isen

0.69

bourg

0.65

omach

0.65

atis

0.64

Activations Density 0.353%

Twitter hashtags or promotional keywords

instances of the letter 'W' and associated patterns in text

No Comments

No Known Activations