INDEX

Explanations

Swedish words

oai_token-act-pair · gpt-3.5-turbo

elements of physical attributes or characteristics

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

IUM

-0.94

untled

-0.93

DragonMagazine

-0.79

TextColor

-0.77

acebook

-0.76

ronics

-0.76

æ©

-0.76

ãĥ¼ãĥĨãĤ£

-0.75

WARE

-0.74

Motor

-0.74

POSITIVE LOGITS

 lang

0.78

los

0.77

tu

0.72

 Esper

0.71

 Nord

0.70

ib

0.68

bre

0.68

vel

0.67

pel

0.67

rh

0.67

Activations Density 0.158%

Swedish words

elements of physical attributes or characteristics

No Comments

No Known Activations