INDEX

Explanations

references to the name "Muhammad"

oai_token-act-pair · gpt-3.5-turbo

references to the name "Muhammad."

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

WD

-0.71

 kindred

-0.70

 Titans

-0.68

ancies

-0.67

 Argon

-0.66

 CentOS

-0.66

 Asheville

-0.65

HMS

-0.65

age

-0.65

aries

-0.65

POSITIVE LOGITS

hammad

1.35

Ali

1.16

iyah

0.95

Ali

0.94

ibn

0.88

 Muhammad

0.84

Yun

0.82

icans

0.80

uddin

0.79

zai

0.79

Activations Density 0.007%

references to the name "Muhammad"

references to the name "Muhammad."

No Comments

No Known Activations