INDEX

Explanations

years expressed in numbers

oai_token-act-pair · gpt-3.5-turbo

numerical data and statistics related to various topics

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

locks

-0.78

staking

-0.77

mble

-0.75

bags

-0.70

gets

-0.69

BOOK

-0.68

pered

-0.64

holders

-0.62

shield

-0.62

bear

-0.61

POSITIVE LOGITS

rd

2.08

RD

1.16

RF

0.94

istics

0.91

osa

0.84

0.82

aquin

0.80

rade

0.80

endez

0.79

enary

0.76

Activations Density 0.091%

years expressed in numbers

numerical data and statistics related to various topics

No Comments

No Known Activations