INDEX

Explanations

ordinal numbers referring to rankings

oai_token-act-pair · gpt-3.5-turbo

numerical rankings and their associated values

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ĵĺ

-0.63

ĸļ

-0.61

 weaving

-0.59

 tirelessly

-0.59

 Mahjong

-0.59

 Daddy

-0.59

 incess

-0.59

 merciless

-0.58

 Anita

-0.56

 hydra

-0.56

POSITIVE LOGITS

th

1.64

rd

1.42

nd

1.37

teenth

1.17

 spots

1.09

st

1.08

ths

1.04

venth

1.03

 points

0.95

ND

0.91

Activations Density 0.081%

ordinal numbers referring to rankings

numerical rankings and their associated values

No Comments

No Known Activations