INDEX

Explanations

political figures and related events

oai_token-act-pair · gpt-3.5-turbo

references to political figures and their interactions

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

zee

-0.79

achine

-0.74

notation

-0.68

abwe

-0.67

zees

-0.67

pta

-0.66

earable

-0.63

haar

-0.63

entin

-0.62

BMC

-0.61

POSITIVE LOGITS

Lago

0.71

 TAMADRA

0.69

fman

0.68

oglu

0.68

BTC

0.63

Os

0.62

 counterparts

0.62

NK

0.61

Invalid

0.61

 counterpart

0.61

Activations Density 0.581%

political figures and related events

references to political figures and their interactions

No Comments

No Known Activations