INDEX

Explanations

phrases that indicate access or permission

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

ieu

-0.80

ernand

-0.79

union

-0.78

boxing

-0.72

ependence

-0.72

aleigh

-0.71

NR

-0.70

resa

-0.69

lication

-0.69

laughter

-0.68

POSITIVE LOGITS

 resources

0.75

 information

0.71

 microphones

0.71

 Pandora

0.70

 trove

0.69

 databases

0.69

 encrypted

0.69

VIP

0.69

 info

0.68

 Whats

0.67

Activations Density 0.029%

phrases that indicate access or permission

No Comments

No Known Activations