INDEX

Explanations

proper nouns, specifically names containing the initial "R"

oai_token-act-pair · gpt-3.5-turbo

occurrences of the letter "R"

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

naire

-0.77

etheless

-0.73

isSpecialOrderable

-0.68

essage

-0.63

ozy

-0.63

icles

-0.61

acion

-0.60

course

-0.60

GMT

-0.59

ocument

-0.59

POSITIVE LOGITS

3.00

1.98

1.40

RI

1.31

RR

1.20

RD

1.19

RA

1.18

RM

1.18

RU

1.18

Rx

1.18

Activations Density 0.037%

proper nouns, specifically names containing the initial "R"

occurrences of the letter "R"

No Comments

No Known Activations