INDEX

Explanations

mentions of the name "Oscar"

oai_token-act-pair · gpt-3.5-turbo

references to the Oscar Awards and related figures

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

ivities

-0.70

ths

-0.69

icago

-0.69

arity

-0.68

£ı

-0.68

INESS

-0.67

ependence

-0.67

 subsistence

-0.67

ipper

-0.66

POSITIVE LOGITS

 Wilde

1.19

 Awards

0.86

 nominations

0.84

 winner

0.80

 nominated

0.75

 Mayer

0.74

 Nebula

0.73

 Award

0.73

 nomination

0.73

WIN

0.72

Activations Density 0.006%

mentions of the name "Oscar"

references to the Oscar Awards and related figures

No Comments

No Known Activations