INDEX

Explanations

proper nouns and trademarked terms

oai_token-act-pair · gpt-3.5-turbo

trademarks or brand indicators

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

glers

-0.95

*/(

-0.81

vernment

-0.77

selage

-0.77

ships

-0.71

 sqor

-0.71

emouth

-0.70

byss

-0.67

bats

-0.67

furt

-0.65

POSITIVE LOGITS

TM

0.95

obile

0.88

ucci

0.84

agic

0.81

asters

0.80

NT

0.79

astic

0.79

ULT

0.78

agnetic

0.77

astics

0.77

Activations Density 0.008%

proper nouns and trademarked terms

trademarks or brand indicators

No Comments

No Known Activations