INDEX

Explanations

proper nouns related to a specific fictional character or family

oai_token-act-pair · gpt-3.5-turbo

references to the character Stark and related names

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

WAYS

-0.79

PER

-0.75

ulhu

-0.74

BLE

-0.72

uters

-0.71

EPA

-0.68

CRE

-0.68

ãĥ¼ãĥ³

-0.68

Ú

-0.67

WHERE

-0.66

POSITIVE LOGITS

 Stark

1.29

 Banner

1.00

iller

0.80

olicy

0.80

auld

0.78

 Lann

0.78

weather

0.75

iffs

0.74

lund

0.72

orman

0.72

Activations Density 0.026%

proper nouns related to a specific fictional character or family

references to the character Stark and related names

No Comments

No Known Activations