INDEX

Explanations

references to "Lord of the Rings"

oai_token-act-pair · gpt-3.5-turbo

references to "The Lord of the Rings" series

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

ï¸ı

-0.65

 puberty

-0.64

OTUS

-0.64

Downloadha

-0.62

inal

-0.61

leader

-0.61

 concentrating

-0.60

 agent

-0.59

Ã§a

-0.58

 partic

-0.58

POSITIVE LOGITS

 Rings

1.20

 Scrolls

0.86

 Mountains

0.84

 Admir

0.80

 Races

0.79

 Mysteries

0.77

 Beasts

0.76

 Balance

0.74

 Seasons

0.73

 Lords

0.73

Activations Density 0.062%

references to "Lord of the Rings"

references to "The Lord of the Rings" series

No Comments

No Known Activations