INDEX

Explanations

phrases indicating interaction or close relationship between multiple entities

oai_token-act-pair · gpt-3.5-turbo

references to interactions or relationships between individuals

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

cribed

-0.60

uci

-0.56

ession

-0.55

Charg

-0.54

Row

-0.54

Hug

-0.53

 Royal

-0.52

 Crusade

-0.51

ãĥ©ãĥ³

-0.51

cluded

-0.51

POSITIVE LOGITS

worldly

0.99

's

0.93

ÃŃs

0.73

é¾įåĸļå£«

0.73

 financially

0.70

 mutually

0.70

opausal

0.68

 verbally

0.68

 sexually

0.67

selves

0.67

Activations Density 0.039%

phrases indicating interaction or close relationship between multiple entities

references to interactions or relationships between individuals

No Comments

No Known Activations