INDEX

Explanations

adjectives related to positive attributes

oai_token-act-pair · gpt-3.5-turbo

words that are phonetically playful or pun-like in nature

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

avia

-0.78

WAR

-0.77

aro

-0.69

ARC

-0.69

raped

-0.69

HI

-0.68

ynthesis

-0.67

chapter

-0.66

 XIII

-0.65

endants

-0.65

POSITIVE LOGITS

ly

1.21

 enough

1.13

ness

1.08

nesses

1.03

est

1.01

LY

0.90

NESS

0.89

 ones

0.87

minded

0.84

liness

0.84

Activations Density 0.523%

adjectives related to positive attributes

words that are phonetically playful or pun-like in nature

No Comments

No Known Activations