INDEX

Explanations

proper nouns, particularly related to specific locations or characters

oai_token-act-pair · gpt-3.5-turbo

names and entities, particularly focusing on the term "Beacon" 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 @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

ivan

-0.77

ktop

-0.76

tery

-0.72

rolet

-0.70

orp

-0.70

est

-0.66

sie

-0.63

avid

-0.62

dos

-0.62

perature

-0.62

POSITIVE LOGITS

 Beacon

0.95

 Triangle

0.74

 exting

0.73

zeb

0.71

ãĤ¤ãĥĪ

0.70

 Kits

0.70

Bom

0.68

 Ridge

0.66

alyst

0.66

bom

0.65

Activations Density 0.013%

proper nouns, particularly related to specific locations or characters

names and entities, particularly focusing on the term "Beacon" and related names

No Comments

No Known Activations