INDEX

Explanations

keywords related to graphs or graphical representations

oai_token-act-pair · gpt-3.5-turbo

references to graphs and graphical representations

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Lauder

-0.65

NESS

-0.62

 Lomb

-0.61

butt

-0.61

 Ambrose

-0.61

ONT

-0.60

estinal

-0.60

 Burton

-0.59

FUL

-0.58

aneous

-0.58

POSITIVE LOGITS

ically

1.29

ical

1.10

ite

1.04

ician

0.88

ql

0.86

ynt

0.85

ophone

0.85

ics

0.85

yx

0.83

icians

0.82

Activations Density 0.035%

keywords related to graphs or graphical representations

references to graphs and graphical representations

No Comments

No Known Activations