INDEX

Explanations

instances of the word "Papal" or related terms within the text

oai_token-act-pair · gpt-3.5-turbo

references to the "Papal" context or related entities

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

nesota

-0.88

IFE

-0.83

hirt

-0.81

é¾

-0.77

ãĤ´ãĥ³

-0.76

pmwiki

-0.74

IGHTS

-0.73

UCT

-0.72

DonaldTrump

-0.72

ACTED

-0.71

POSITIVE LOGITS

yrus

1.29

acy

1.03

arella

0.96

ular

0.93

rika

0.90

olitan

0.90

erson

0.89

ros

0.89

ropri

0.89

azz

0.87

Activations Density 0.015%

instances of the word "Papal" or related terms within the text

references to the "Papal" context or related entities

No Comments

No Known Activations