INDEX

Explanations

names or terms related to the name "Sharon"

oai_token-act-pair · gpt-3.5-turbo

mentions of a specific individual or entity named "Shar."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

REDACTED

-0.71

xual

-0.65

-0.64

es

-0.63

 Titanic

-0.62

 Uran

-0.59

 interchange

-0.58

 accur

-0.58

é»Ĵ

-0.57

 Hawaiian

-0.57

POSITIVE LOGITS

pton

1.44

pless

1.21

ples

1.16

ply

1.16

pling

1.11

jah

1.05

pe

1.02

riors

1.00

ps

1.00

raq

0.96

Activations Density 0.021%

names or terms related to the name "Sharon"

mentions of a specific individual or entity named "Shar."

No Comments

No Known Activations