INDEX

Explanations

the letter 'n' followed by a number

oai_token-act-pair · gpt-3.5-turbo

instances of the letter 'n' appearing 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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Remastered

-0.64

 frank

-0.62

 decomp

-0.59

 WRITE

-0.58

Vaj

-0.57

spr

-0.57

 slack

-0.56

 Schr

-0.55

 Presidency

-0.55

 polyg

-0.55

POSITIVE LOGITS

autical

1.45

ameless

1.44

uggets

1.39

ectar

1.37

ativity

1.34

imble

1.31

ikuman

1.25

ymph

1.21

anny

1.20

atures

1.20

Activations Density 0.020%

the letter 'n' followed by a number

instances of the letter 'n' appearing in various contexts

No Comments

No Known Activations