INDEX

Explanations

adverbs used to convey intensity or emphasis

oai_token-act-pair · gpt-3.5-turbo

expressions related to contemplation and evaluation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

adra

-0.79

catentry

-0.71

ante

-0.66

edition

-0.64

ario

-0.63

%"

-0.61

é»Ĵ

-0.60

ource

-0.60

endor

-0.58

riott

-0.58

POSITIVE LOGITS

 Bout

0.74

 unthinkable

0.66

 provoking

0.66

oby

0.64

 aloud

0.64

ti

0.64

cho

0.63

mas

0.63

orth

0.62

esian

0.62

Activations Density 0.132%

adverbs used to convey intensity or emphasis

expressions related to contemplation and evaluation

No Comments

No Known Activations