INDEX

Explanations

information or discussions about the inner workings of various systems or organizations

oai_token-act-pair · gpt-3.5-turbo

references to internal mechanisms or systems

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

atoes

-0.83

eday

-0.81

essors

-0.80

enegger

-0.79

oulos

-0.78

orthy

-0.77

enance

-0.76

HAHAHAHA

-0.72

abis

-0.69

DragonMagazine

-0.68

POSITIVE LOGITS

 workings

1.22

most

1.16

 sanct

0.85

 combustion

0.79

ranean

0.77

jection

0.76

 circle

0.75

verted

0.73

 compass

0.72

 essence

0.71

Activations Density 0.020%

information or discussions about the inner workings of various systems or organizations

references to internal mechanisms or systems

No Comments

No Known Activations