INDEX

Explanations

specific names, such as "Miller", "Gustav", "Troutman", and "Vaughn"

oai_token-act-pair · gpt-3.5-turbo

specific names and identifiers related to people, locations, and organizations

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

entle

-0.51

rill

-0.49

reau

-0.48

ween

-0.47

estival

-0.47

eteria

-0.47

ovies

-0.46

MU

-0.46

 vegetables

-0.45

 welf

-0.45

POSITIVE LOGITS

ciation

0.60

Io

0.58

bilt

0.53

 Saturn

0.53

imaru

0.53

Zot

0.53

ãĥĺãĥ©

0.50

 Polish

0.49

 Cerberus

0.48

ģ«

0.48

Activations Density 1.466%

specific names, such as "Miller", "Gustav", "Troutman", and "Vaughn"

specific names and identifiers related to people, locations, and organizations

No Comments

No Known Activations