INDEX

Explanations

names of personalities or public figures

oai_token-act-pair · gpt-3.5-turbo

vowel-heavy words

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

Reloaded

-0.80

 srfAttach

-0.77

CLASSIFIED

-0.74

ADE

-0.71

DERR

-0.70

Sharp

-0.69

 withd

-0.67

ãĥķãĤ©

-0.66

ENDED

-0.66

ãĥ¯ãĥ³

-0.64

POSITIVE LOGITS

ghan

1.09

pless

0.99

ghai

0.96

pling

0.91

vel

0.91

lder

0.90

vern

0.90

ju

0.88

veland

0.86

ze

0.86

Activations Density 0.121%

names of personalities or public figures

vowel-heavy words

No Comments

No Known Activations