INDEX

Explanations

phrases indicating tension or conflict between different entities

oai_token-act-pair · gpt-3.5-turbo

phrases that discuss relationships or tensions between entities, particularly between countries or groups

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

nit

-0.92

unc

-0.82

bye

-0.82

OGR

-0.79

ãĤ®

-0.76

gow

-0.75

inventoryQuantity

-0.74

TPPStreamerBot

-0.73

nell

-0.72

daq

-0.72

POSITIVE LOGITS

 sexes

1.29

 genders

1.16

two

0.96

 combatants

0.94

 halves

0.93

 factions

0.92

 sides

0.90

 parties

0.88

 spouses

0.88

 them

0.86

Activations Density 0.063%

phrases indicating tension or conflict between different entities

phrases that discuss relationships or tensions between entities, particularly between countries or groups

No Comments

No Known Activations