INDEX

Explanations

disambiguation pages indicating different meanings of terms or names

oai_token-act-pair · gpt-3.5-turbo

references to multiple uses or definitions of a term

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

WAY

-0.65

EMA

-0.65

milo

-0.63

aturdays

-0.63

MN

-0.63

LU

-0.60

CHR

-0.59

ufact

-0.59

KA

-0.58

Zen

-0.58

POSITIVE LOGITS

pmwiki

0.94

wiki

0.79

 redirect

0.77

 Citiz

0.77

ipedia

0.77

trop

0.75

species

0.73

wik

0.73

".[

0.73

 Wiki

0.72

Activations Density 0.138%

disambiguation pages indicating different meanings of terms or names

references to multiple uses or definitions of a term

No Comments

No Known Activations