INDEX

Explanations

specific points or moments in time

oai_token-act-pair · gpt-3.5-turbo

repeated phrases indicating points in time

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

cause

-0.70

mor

-0.66

Cause

-0.63

 Himself

-0.61

ums

-0.61

MpServer

-0.59

dr

-0.58

umb

-0.58

anium

-0.58

uld

-0.57

POSITIVE LOGITS

we

0.95

 they

0.91

 there

0.90

it

0.75

he

0.74

 however

0.70

 pandemonium

0.70

she

0.69

0.66

adays

0.66

Activations Density 0.248%

specific points or moments in time

repeated phrases indicating points in time

No Comments

No Known Activations