INDEX

Explanations

mentions of large things or entities

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "large" and related terms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

vir

-0.85

BIL

-0.71

 Dmitry

-0.71

fell

-0.71

Luke

-0.70

undown

-0.70

Ark

-0.70

Putin

-0.69

fly

-0.68

Ukraine

-0.68

POSITIVE LOGITS

arge

2.50

 Franc

1.68

 Mund

1.47

 Glad

1.43

 Pats

1.32

 Madd

1.29

 franc

1.18

agues

1.08

 panc

1.07

 Joyce

1.04

Activations Density 0.043%

mentions of large things or entities

occurrences of the word "large" and related terms

No Comments

No Known Activations