INDEX

Explanations

references to the Academy Awards

oai_token-act-pair · gpt-3.5-turbo

references to the Academy Awards

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

 Meier

-0.76

TPS

-0.73

ppings

-0.69

pd

-0.69

bare

-0.68

 Bundy

-0.68

ï¸ı

-0.67

OPLE

-0.66

ulkan

-0.66

drivers

-0.65

POSITIVE LOGITS

emies

1.07

 Awards

1.03

ritic

0.87

 Ruins

0.83

 Award

0.80

uates

0.74

alon

0.74

redit

0.73

rylic

0.70

Grade

0.70

Activations Density 0.022%

references to the Academy Awards

references to the Academy Awards

No Comments

No Known Activations