INDEX

Explanations

statements or phrases related to transparency or openness

oai_token-act-pair · gpt-3.5-turbo

instances of the word "open" in various contexts related to openness and accessibility

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

iage

-0.75

ILA

-0.69

CTV

-0.67

grade

-0.66

ILL

-0.64

ogun

-0.64

doms

-0.64

ufact

-0.63

ibur

-0.63

IVERS

-0.63

POSITIVE LOGITS

 minded

0.80

open

0.79

office

0.79

 spaces

0.77

 wounds

0.75

 sidel

0.72

ended

0.72

bies

0.70

 open

0.70

 doors

0.69

Activations Density 0.035%

statements or phrases related to transparency or openness

instances of the word "open" in various contexts related to openness and accessibility

No Comments

No Known Activations