INDEX

Explanations

words related to wars and conflicts

oai_token-act-pair · gpt-3.5-turbo

references to specific events or individuals

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

Newsletter

-0.86

aukee

-0.76

aucus

-0.70

nyder

-0.66

 Legislature

-0.65

dozen

-0.65

colored

-0.64

CLOSE

-0.63

LOS

-0.62

ASHINGTON

-0.60

POSITIVE LOGITS

 Whilst

1.28

 whilst

1.18

 util

1.04

ilst

0.94

 organisers

0.90

 emphas

0.88

 CCTV

0.84

 organisations

0.83

 criticised

0.81

 realise

0.80

Activations Density 1.499%

words related to wars and conflicts

references to specific events or individuals

No Comments

No Known Activations