INDEX

Explanations

proper nouns

oai_token-act-pair · gpt-3.5-turbo

mentions of the individual named Watts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

istically

-0.72

oral

-0.70

ournal

-0.67

ptive

-0.66

esan

-0.65

amon

-0.64

oslov

-0.63

 Lyft

-0.63

oin

-0.62

viously

-0.62

POSITIVE LOGITS

atts

1.24

 Watts

1.12

hots

0.86

haw

0.84

acre

0.82

 Compton

0.80

cliffe

0.80

aga

0.79

ELF

0.79

kins

0.78

Activations Density 0.006%

proper nouns

mentions of the individual named Watts

No Comments

No Known Activations