INDEX

Explanations

phrases regarding legal charges and offenses

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

 obser

-0.74

 contrace

-0.71

soDeliveryDate

-0.70

tnc

-0.69

 Voices

-0.68

 Generations

-0.67

ciating

-0.65

natureconservancy

-0.65

 Nobel

-0.64

 landscapes

-0.64

POSITIVE LOGITS

 punishable

1.08

 offences

1.06

 misdemeanor

1.03

 felony

1.02

 misdem

1.02

 offenses

1.01

 criminal

0.97

criminal

0.96

 indict

0.96

 alleging

0.95

Activations Density 0.009%

phrases regarding legal charges and offenses

No Comments

No Known Activations