INDEX

Explanations

mentions of the word "Queen"

oai_token-act-pair · gpt-3.5-turbo

mentions of royalty, specifically the term "Queen."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

mental

-0.80

odcast

-0.79

riks

-0.72

andem

-0.68

urgical

-0.67

chell

-0.66

kson

-0.66

trak

-0.65

ãĤ·ãĥ£

-0.63

ript

-0.62

POSITIVE LOGITS

 Anne

1.07

Queen

1.03

 Queen

1.01

stown

0.93

esses

0.82

Anne

0.82

 Mother

0.81

 Bride

0.80

Bee

0.80

 Mary

0.76

Activations Density 0.006%

mentions of the word "Queen"

mentions of royalty, specifically the term "Queen."

No Comments

No Known Activations