INDEX

Explanations

expletives or strong vulgar language

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

Ĭ±

-0.73

 ancest

-0.71

 mosqu

-0.70

 pestic

-0.67

utterstock

-0.67

æĪ¦

-0.66

 rece

-0.66

 reluct

-0.64

Decre

-0.64

 reper

-0.63

POSITIVE LOGITS

hole

1.18

holes

1.15

tty

1.02

king

0.98

cking

0.94

kers

0.92

shit

0.88

gger

0.88

fuck

0.85

ing

0.85

Activations Density 0.009%

expletives or strong vulgar language

No Comments

No Known Activations