INDEX

Explanations

words related to obedience and disobedience in the context of social or personal dynamics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs49152-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

49,152

Data Type

torch.float32

Hook Point

blocks.8.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

 Lauder

-0.83

 Divinity

-0.79

UAL

-0.78

 Painter

-0.77

ILLE

-0.75

âĸ¬

-0.71

 Birch

-0.69

UES

-0.68

EngineDebug

-0.68

 Martial

-0.68

POSITIVE LOGITS

vious

1.28

lig

1.23

served

1.23

esity

1.21

amacare

1.18

fusc

1.14

serv

1.13

lique

1.12

server

1.11

acter

1.11

Activations Density 0.008%

words related to obedience and disobedience in the context of social or personal dynamics

No Comments

No Known Activations