INDEX

Explanations

phrases indicating a period leading up to an event or occurrence

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

inav

-0.83

kan

-0.76

icular

-0.69

sf

-0.66

oos

-0.64

©¶æ

-0.63

owsky

-0.62

 conserv

-0.61

tle

-0.61

mes

-0.61

POSITIVE LOGITS

 thereto

0.91

uberty

0.76

actionDate

0.75

ounter

0.72

isode

0.72

stage

0.71

dating

0.71

imester

0.69

 Election

0.68

mortem

0.67

Activations Density 0.089%

phrases indicating a period leading up to an event or occurrence

No Comments

No Known Activations