INDEX

Explanations

proper nouns or acronyms beginning with the letters "TR"

oai_token-act-pair · gpt-3.5-turbo

references to a specific entity or topic denoted by "TR"

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

manship

-0.92

soever

-0.78

lace

-0.75

âĸ¬âĸ¬

-0.69

 arts

-0.69

creen

-0.68

 Opera

-0.67

bats

-0.67

gaard

-0.66

eers

-0.66

POSITIVE LOGITS

UTH

1.26

ACK

1.07

UST

0.98

UCK

0.97

ACTION

0.93

IP

0.93

umble

0.92

ickle

0.91

ACT

0.91

ANS

0.90

Activations Density 0.017%

proper nouns or acronyms beginning with the letters "TR"

references to a specific entity or topic denoted by "TR"

No Comments

No Known Activations