INDEX

Explanations

comparative statements indicating a quantity greater than

oai_token-act-pair · gpt-3.5-turbo

comparative phrases indicating quantities or numbers

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

EMENT

-0.69

CLASSIFIED

-0.66

imity

-0.65

URR

-0.65

 Juda

-0.64

ALE

-0.62

ALSE

-0.60

ICT

-0.60

ilic

-0.59

Alz

-0.59

POSITIVE LOGITS

whelming

0.76

 usual

0.76

atos

0.75

drive

0.67

than

0.67

amaru

0.64

usual

0.63

whel

0.62

pload

0.60

integer

0.60

Activations Density 0.091%

comparative statements indicating a quantity greater than

comparative phrases indicating quantities or numbers

No Comments

No Known Activations