INDEX

Explanations

names of notable individuals or significant figures

oai_token-act-pair · gpt-3.5-turbo

names of individuals mentioned in the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

 tune

-0.68

USC

-0.65

 chem

-0.65

 womb

-0.61

 Hurricane

-0.61

CTR

-0.59

 Spectre

-0.58

NRS

-0.58

 Mandatory

-0.58

TBA

-0.57

POSITIVE LOGITS

 apologised

1.03

itsch

0.89

yip

0.88

cott

0.87

zinski

0.87

ĳå£«

0.87

opoulos

0.84

iewicz

0.84

arde

0.83

ovich

0.82

Activations Density 0.134%

names of notable individuals or significant figures

names of individuals mentioned in the text

No Comments

No Known Activations