INDEX

Explanations

words related to direct and straightforward communication

oai_token-act-pair · gpt-3.5-turbo

the word "blunt" and its variations

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

Haunted

-0.79

osponsors

-0.78

Lv

-0.72

Sync

-0.72

ammy

-0.71

HB

-0.70

alian

-0.69

YL

-0.68

FactoryReloaded

-0.68

————————

-0.68

POSITIVE LOGITS

 blunt

1.12

odox

0.80

 arrow

0.79

ball

0.78

sonian

0.76

 strokes

0.74

wit

0.74

 instrument

0.73

tip

0.72

 skew

0.72

Activations Density 0.012%

words related to direct and straightforward communication

the word "blunt" and its variations

No Comments

No Known Activations