INDEX

Explanations

addresses or locations in text

oai_token-act-pair · gpt-3.5-turbo

specific addresses referenced in a context related to location or communication

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

uj

-0.78

apt

-0.75

lime

-0.69

ipeg

-0.68

ooks

-0.68

ischer

-0.68

pload

-0.67

drm

-0.67

 Kirin

-0.66

rd

-0.66

POSITIVE LOGITS

 addr

0.85

holder

0.83

holders

0.81

 Address

0.80

 address

0.79

onica

0.71

 addresses

0.69

sterdam

0.66

Address

0.66

nikov

0.65

Activations Density 0.018%

addresses or locations in text

specific addresses referenced in a context related to location or communication

No Comments

No Known Activations