INDEX

Explanations

possession or ownership references, such as "of their own"

oai_token-act-pair · gpt-3.5-turbo

sequences related to possession and ownership concepts

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

 redacted

-0.49

ngth

-0.48

anwhile

-0.48

 inval

-0.48

ources

-0.47

onest

-0.46

 cropped

-0.46

pse

-0.45

reau

-0.45

 Loaded

-0.45

POSITIVE LOGITS

 apocalypse

0.53

IPM

0.51

 convergence

0.49

 modification

0.48

venge

0.47

externalActionCode

0.47

 ACTIONS

0.47

Council

0.47

 renaissance

0.47

 reconciliation

0.46

Activations Density 1.390%

possession or ownership references, such as "of their own"

sequences related to possession and ownership concepts

No Comments

No Known Activations