INDEX

Explanations

years, numbers, and statistics

oai_token-act-pair · gpt-3.5-turbo

references to specific years and numerical data

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

iday

-0.77

iped

-0.74

PW

-0.72

ition

-0.71

ishing

-0.71

ingham

-0.71

igating

-0.70

ential

-0.68

 Palest

-0.67

ary

-0.66

POSITIVE LOGITS

teenth

1.05

 satell

0.91

 millenn

0.87

UTH

0.78

0.77

ptive

0.76

veyard

0.74

eteenth

0.73

antle

0.73

otomy

0.72

Activations Density 0.055%

years, numbers, and statistics

references to specific years and numerical data

No Comments

No Known Activations