INDEX

Explanations

mentions of the name "Lee"

oai_token-act-pair · gpt-3.5-turbo

occurrences of the substring "ee" in various contexts

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

 tert

-0.69

iques

-0.68

urgical

-0.66

izoph

-0.64

 honesty

-0.62

CPC

-0.61

enegger

-0.60

atform

-0.60

ĺħ

-0.59

icity

-0.58

POSITIVE LOGITS

pee

1.41

ples

1.32

gee

1.27

zy

1.25

ves

1.23

pees

1.22

bles

1.22

zer

1.21

bies

1.16

bee

1.13

Activations Density 0.059%

mentions of the name "Lee"

occurrences of the substring "ee" in various contexts

No Comments

No Known Activations