INDEX

Explanations

proper nouns related to characters or names

oai_token-act-pair · gpt-3.5-turbo

characters and their relationships within a narrative

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 newsletters

-0.84

Islamic

-0.78

India

-0.77

Japan

-0.76

liga

-0.75

utra

-0.72

 PRESS

-0.69

intage

-0.67

iments

-0.67

umbai

-0.65

POSITIVE LOGITS

 Luthor

1.06

 Crusher

0.97

 Reyes

0.95

hyde

0.95

 Finch

0.94

Osw

0.91

 Stark

0.90

 Locke

0.90

 Winchester

0.90

 Sawyer

0.89

Activations Density 0.254%

proper nouns related to characters or names

characters and their relationships within a narrative

No Comments

No Known Activations