INDEX

Explanations

references to the city of Toronto

oai_token-act-pair · gpt-3.5-turbo

mentions of the term "TOR" and variations associated with Toronto-related content or technology

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

ãħĭ

-0.71

scl

-0.70

wagen

-0.70

holder

-0.69

essor

-0.67

 WARN

-0.65

 Nationwide

-0.63

sam

-0.63

forces

-0.62

Doe

-0.62

POSITIVE LOGITS

onto

1.19

mented

0.98

rent

0.84

oshenko

0.83

TOR

0.81

MENT

0.81

heed

0.80

othy

0.79

rance

0.77

rence

0.76

Activations Density 0.012%

references to the city of Toronto

mentions of the term "TOR" and variations associated with Toronto-related content or technology

No Comments

No Known Activations