INDEX

Explanations

phrases related to modifications or changes

oai_token-act-pair · gpt-3.5-turbo

words related to actions of restoring, disassembling, or transitioning between states

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

 Sierra

-0.71

 formation

-0.63

 spont

-0.62

lihood

-0.62

HOME

-0.59

 divergence

-0.59

 attitudes

-0.58

 sake

-0.57

 ability

-0.57

 impulse

-0.57

POSITIVE LOGITS

rolled

1.12

inged

1.12

idered

1.05

arily

1.03

fed

1.02

alysed

1.01

ioned

0.99

arded

0.97

ogged

0.97

icked

0.95

Activations Density 0.290%

phrases related to modifications or changes

words related to actions of restoring, disassembling, or transitioning between states

No Comments

No Known Activations