INDEX

Explanations

names of individuals or last names

oai_token-act-pair · gpt-3.5-turbo

names related to individuals involved in leadership or notable events

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

aries

-0.78

ARY

-0.75

ocal

-0.74

eral

-0.74

arist

-0.73

glers

-0.70

ptions

-0.69

artz

-0.67

ãĥ´

-0.66

axy

-0.66

POSITIVE LOGITS

 ende

0.86

DragonMagazine

0.83

£ı

0.82

NetMessage

0.78

 flourish

0.77

ly

0.72

 Patterson

0.70

Ħ¢

0.69

etics

0.67

endale

0.66

Activations Density 0.025%

names of individuals or last names

names related to individuals involved in leadership or notable events

No Comments

No Known Activations