INDEX

Explanations

mentions of the Catholic community

oai_token-act-pair · gpt-3.5-turbo

repeated references to the Catholic faith and its followers

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

yss

-0.68

inventoryQuantity

-0.66

jri

-0.65

ļé

-0.60

series

-0.58

sa

-0.58

sole

-0.57

oÄŁ

-0.57

 nerve

-0.56

 administration

-0.56

POSITIVE LOGITS

aurus

1.01

ervatives

0.95

ervative

0.93

 Catholics

0.91

cius

0.88

hip

0.83

paces

0.78

hips

0.75

 rejoice

0.72

alam

0.72

Activations Density 0.018%

mentions of the Catholic community

repeated references to the Catholic faith and its followers

No Comments

No Known Activations