INDEX

Explanations

phrases related to military conflict

oai_token-act-pair · gpt-3.5-turbo

references to war

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

essee

-0.83

FORMATION

-0.77

aminer

-0.73

ostics

-0.70

Ħ¢

-0.70

 Universe

-0.69

prints

-0.69

Distance

-0.69

sembly

-0.69

gow

-0.67

POSITIVE LOGITS

fare

1.22

lords

1.19

lord

1.17

riors

1.12

rior

1.07

fighting

0.99

 waged

0.97

fighter

0.92

ring

0.90

zone

0.84

Activations Density 0.033%

phrases related to military conflict

references to war

No Comments

No Known Activations