INDEX

Explanations

specific symbols and characters in a text

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "forbid" and its variations, alongside specific symbols and abbreviations

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

uations

-0.75

urally

-0.74

uation

-0.73

orem

-0.72

eur

-0.71

emouth

-0.70

heed

-0.68

owan

-0.66

ciating

-0.65

ements

-0.65

POSITIVE LOGITS

ļéĨĴ

0.93

bie

0.87

atri

0.87

earance

0.87

stract

0.84

é¾įåĸļå£«

0.74

bian

0.74

icity

0.72

edia

0.71

Reply

0.71

Activations Density 0.051%

specific symbols and characters in a text

occurrences of the word "forbid" and its variations, alongside specific symbols and abbreviations

No Comments

No Known Activations