INDEX

Explanations

positive traits or characteristics about individuals or entities

oai_token-act-pair · gpt-3.5-turbo

positive adjectives and their intensifiers

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

Ħ¢

-0.80

early

-0.74

jen

-0.73

POST

-0.71

hower

-0.71

advertising

-0.71

merce

-0.70

ī

-0.70

late

-0.69

steps

-0.67

POSITIVE LOGITS

 knack

0.93

 chance

0.93

 credibility

0.79

 opportunity

0.76

 legs

0.76

 ambitions

0.75

 flair

0.75

 flexibility

0.74

 penchant

0.73

 advantages

0.73

Activations Density 0.315%

positive traits or characteristics about individuals or entities

positive adjectives and their intensifiers

No Comments

No Known Activations