INDEX

Explanations

proper names

oai_token-act-pair · gpt-3.5-turbo

mentions of a specific individual named Ke

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

etheless

-0.82

ashtra

-0.75

schild

-0.74

 tremend

-0.70

rition

-0.70

sburgh

-0.69

åħī

-0.69

raints

-0.68

ancial

-0.68

 guiActiveUnfocused

-0.68

POSITIVE LOGITS

ptic

1.03

Ke

1.03

SPA

0.99

eling

0.98

lem

0.95

cker

0.95

gan

0.95

aton

0.94

isha

0.92

bye

0.92

Activations Density 0.004%

proper names

mentions of a specific individual named Ke

No Comments

No Known Activations