INDEX

Explanations

adverbs that indicate degree or intensity

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

ĺħ

-0.78

eday

-0.78

ODY

-0.73

 Tags

-0.68

OGR

-0.68

selage

-0.67

 Korea

-0.66

arella

-0.66

ydia

-0.65

OTOS

-0.65

POSITIVE LOGITS

 divisive

0.86

 lethal

0.81

 destructive

0.81

 chaotic

0.81

 controversial

0.80

 mythical

0.80

 abras

0.80

 embattled

0.80

 politic

0.80

 infamous

0.80

Activations Density 0.077%

adverbs that indicate degree or intensity

No Comments

No Known Activations