INDEX

Explanations

the word "leaf" and variations of it

oai_token-act-pair · gpt-3.5-turbo

references to leaves and related botanical terms

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

PDATE

-0.81

nels

-0.73

 Revel

-0.65

 attackers

-0.63

cedes

-0.63

 indemn

-0.61

 Doyle

-0.60

AUD

-0.60

AAF

-0.60

ICAN

-0.60

POSITIVE LOGITS

let

1.37

lets

1.21

leted

1.08

leaf

1.02

lete

0.94

iard

0.92

ho

0.88

letes

0.87

 leaf

0.87

doms

0.86

Activations Density 0.052%

the word "leaf" and variations of it

references to leaves and related botanical terms

No Known Activations