INDEX

Explanations

specific nouns and terms related to various topics such as animals, politics, technology, and organizations

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

Interstitial

-0.73

Ĥİ

-0.66

cffffcc

-0.64

 Leban

-0.63

steamapps

-0.62

athered

-0.61

eping

-0.61

NetMessage

-0.60

YING

-0.60

nova

-0.59

POSITIVE LOGITS

soever

0.96

 corresponds

0.88

 they

0.84

he

0.77

we

0.75

you

0.72

 THEY

0.72

yip

0.71

 qualifies

0.71

 constitutes

0.70

Activations Density 0.054%

specific nouns and terms related to various topics such as animals, politics, technology, and organizations

No Comments

No Known Activations