INDEX

Explanations

popular items or concepts

oai_token-act-pair · gpt-3.5-turbo

references to the concept of popularity

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

ignt

-0.75

ASC

-0.70

ĸļ

-0.67

agher

-0.67

ural

-0.67

lean

-0.65

omething

-0.65

RAW

-0.64

apo

-0.64

ERO

-0.64

POSITIVE LOGITS

ized

1.17

ised

1.04

izing

1.00

ity

0.98

ly

0.94

ize

0.90

 tourist

0.86

izers

0.85

 destinations

0.84

ization

0.83

Activations Density 0.044%

popular items or concepts

references to the concept of popularity

No Comments

No Known Activations