INDEX

Explanations

references to places of detention, particularly jails

oai_token-act-pair · gpt-3.5-turbo

references to jails or imprisonment

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

 Nare

-0.79

FORE

-0.73

yip

-0.73

 Scotia

-0.72

 Bundes

-0.69

xit

-0.67

chell

-0.65

 livest

-0.63

ndum

-0.63

ilater

-0.63

POSITIVE LOGITS

breaks

0.85

broken

0.85

break

0.85

strap

0.85

rooms

0.84

house

0.82

breakers

0.80

 jails

0.78

itory

0.77

 jail

0.76

Activations Density 0.019%

references to places of detention, particularly jails

references to jails or imprisonment

No Comments

No Known Activations