INDEX

Explanations

phrases or words related to location or position

oai_token-act-pair · gpt-3.5-turbo

phrases indicating location or presence in time and space

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

 Logged

-0.85

STA

-0.76

MU

-0.67

 PsyNet

-0.67

 Proxy

-0.64

 Advertisement

-0.63

BS

-0.62

FB

-0.62

ANN

-0.62

======

-0.61

POSITIVE LOGITS

 periphery

0.91

yond

0.72

 contiguous

0.70

ibaba

0.70

 vicinity

0.70

between

0.69

abouts

0.68

 coasts

0.68

 continuum

0.66

 Isles

0.66

Activations Density 0.128%

phrases or words related to location or position

phrases indicating location or presence in time and space

No Comments

No Known Activations