INDEX

Explanations

references to numerical values followed by the word "man"

oai_token-act-pair · gpt-3.5-turbo

the term "man" in various contexts

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

¥ŀ

-0.75

PsyNetMessage

-0.74

 Blackwell

-0.73

uesday

-0.67

lease

-0.65

Ire

-0.64

CAST

-0.63

ĸļ

-0.62

avorite

-0.61

 Beir

-0.60

POSITIVE LOGITS

ifest

1.45

uscript

1.18

ufact

1.05

iac

1.02

hood

1.01

hunt

1.01

ned

1.00

WithNo

0.99

oeuv

0.97

ning

0.95

Activations Density 0.031%

references to numerical values followed by the word "man"

the term "man" in various contexts

No Comments

No Known Activations