INDEX

Explanations

questions or prompts within the text

oai_token-act-pair · gpt-3.5-turbo

questions posed within the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

roud

-0.66

ographed

-0.63

estial

-0.55

hap

-0.53

æĪ¦

-0.51

itton

-0.50

apsed

-0.49

 tant

-0.49

ICT

-0.49

rib

-0.48

POSITIVE LOGITS

Why

1.35

why

1.27

Why

1.25

why

1.24

WHY

1.22

 Does

1.22

How

1.18

Does

1.17

Who

1.13

 What

1.12

Activations Density 0.146%

questions or prompts within the text

questions posed within the text

No Comments

No Known Activations