INDEX

Explanations

terms related to "Gr" followed by a number

oai_token-act-pair · gpt-3.5-turbo

occurrences of the prefix "Gr."

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

eers

-0.73

quo

-0.65

Lei

-0.64

eer

-0.63

 scouts

-0.63

 writ

-0.62

manship

-0.62

 Scouting

-0.62

 foul

-0.60

 reserved

-0.59

POSITIVE LOGITS

iffin

1.33

anny

1.31

umpy

1.29

anted

1.27

inding

1.25

udge

1.25

itty

1.20

ayson

1.19

aced

1.16

acies

1.16

Activations Density 0.020%

terms related to "Gr" followed by a number

occurrences of the prefix "Gr."

No Comments

No Known Activations