INDEX

Explanations

words related to locks and locking devices

oai_token-act-pair · gpt-3.5-turbo

terms related to locking mechanisms or security systems

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

issance

-0.72

ional

-0.70

olf

-0.63

OTAL

-0.62

 constitu

-0.62

ial

-0.61

schild

-0.60

ordan

-0.58

 Legion

-0.57

 Winc

-0.57

POSITIVE LOGITS

picking

1.38

heed

1.23

pick

1.16

step

1.14

down

0.99

mith

0.99

creen

0.98

downs

0.98

lear

0.97

hart

0.97

Activations Density 0.042%

words related to locks and locking devices

terms related to locking mechanisms or security systems

No Comments

No Known Activations