INDEX

Explanations

words ending in 'lee'

oai_token-act-pair · gpt-3.5-turbo

the name "Lee" 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 @ 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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

SOC

-0.75

 Compass

-0.73

 Regions

-0.66

 bore

-0.62

lib

-0.62

ãĥĳ

-0.61

 nitrogen

-0.60

 Atlas

-0.59

 qualification

-0.59

 manifests

-0.58

POSITIVE LOGITS

lee

4.85

lees

2.92

leeve

2.01

leigh

1.42

Lee

1.27

leys

1.23

lynn

1.22

kee

1.15

lie

1.12

ley

1.12

Activations Density 0.016%

words ending in 'lee'

the name "Lee" in various contexts

No Comments

No Known Activations