INDEX

Explanations

names of individuals related to politics and news headlines

oai_token-act-pair · gpt-3.5-turbo

mentions of specific individuals, particularly in relation to statements or reports

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

 deaf

-0.70

orian

-0.67

Lex

-0.67

Mart

-0.66

 Occupations

-0.66

essors

-0.66

teenth

-0.65

Wars

-0.65

FUL

-0.63

laus

-0.63

POSITIVE LOGITS

 Weiner

1.12

 Abedin

0.96

aukee

0.87

ufact

0.74

arty

0.74

 needles

0.73

ullivan

0.72

ãĤ¦ãĤ¹

0.71

agog

0.70

combe

0.69

Activations Density 0.006%

names of individuals related to politics and news headlines

mentions of specific individuals, particularly in relation to statements or reports

No Comments

No Known Activations