INDEX

Explanations

references to the act of dressing or clothing

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

FTA

-0.76

é¾įå¥ĳå£«

-0.73

rals

-0.70

ntil

-0.69

ategory

-0.69

ropy

-0.69

JV

-0.68

ecd

-0.66

pelling

-0.65

fram

-0.65

POSITIVE LOGITS

 rooms

1.01

 gown

0.98

rooms

0.94

room

0.94

 Room

0.86

 Sands

0.83

 room

0.82

apore

0.80

ments

0.80

mond

0.78

Activations Density 0.003%

references to the act of dressing or clothing

No Comments

No Known Activations