INDEX

Explanations

names and terms related to specific individuals

oai_token-act-pair · gpt-3.5-turbo

specific names, particularly of individuals or groups associated with notable statements or actions

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

imb

-0.68

onomy

-0.67

bidden

-0.67

aceous

-0.66

landish

-0.66

psons

-0.65

ware

-0.65

åĭ

-0.64

ĨĴ

-0.63

phabet

-0.63

POSITIVE LOGITS

ggies

0.78

agher

0.71

isite

0.70

 Hussain

0.70

---------------

0.70

asar

0.68

 Mush

0.67

rooms

0.67

idence

0.65

rences

0.64

Activations Density 0.042%

names and terms related to specific individuals

specific names, particularly of individuals or groups associated with notable statements or actions

No Comments

No Known Activations