INDEX

Explanations

the word 'cle' combined with varying sequences of letters or symbols

oai_token-act-pair · gpt-3.5-turbo

the prefix "cle" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Ducks

-0.77

ãĥ¼ãĥĨãĤ£

-0.67

MFT

-0.64

 Krug

-0.64

 Shogun

-0.64

yrinth

-0.64

BuyableInstoreAndOnline

-0.63

 Galaxy

-0.63

Tul

-0.63

 Panther

-0.62

POSITIVE LOGITS

arest

1.16

avage

1.05

veland

0.98

combe

0.92

utic

0.92

aving

0.92

ipe

0.92

asp

0.91

ician

0.87

iff

0.84

Activations Density 0.010%

the word 'cle' combined with varying sequences of letters or symbols

the prefix "cle" in various contexts

No Comments

No Known Activations