INDEX

Explanations

references to things being empty

oai_token-act-pair · gpt-3.5-turbo

mentions of the word "empty."

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.5.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

abol

-0.84

irtual

-0.72

arya

-0.71

Murray

-0.70

 appropri

-0.69

ection

-0.69

NEWS

-0.69

CVE

-0.69

ect

-0.69

dar

-0.67

POSITIVE LOGITS

 empty

0.86

empty

0.80

 space

0.79

 Empty

0.79

 spaces

0.78

igue

0.77

 shells

0.77

 vacancies

0.75

 slate

0.74

 bottles

0.73

Activations Density 0.016%

references to things being empty

mentions of the word "empty."

No Comments

No Known Activations