INDEX

Explanations

numerical values indicating emphasis or importance

oai_token-act-pair · gpt-3.5-turbo

frequent occurrences of the word "Even"

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

cel

-0.69

ode

-0.68

acht

-0.65

rend

-0.64

ãĤ¿

-0.64

eding

-0.63

RFC

-0.63

ocyte

-0.62

ASE

-0.60

osal

-0.60

POSITIVE LOGITS

 though

0.78

more

0.76

 remotely

0.73

Berry

0.67

antine

0.67

withstanding

0.66

song

0.66

ettes

0.65

ront

0.65

ivari

0.65

Activations Density 0.014%

numerical values indicating emphasis or importance

frequent occurrences of the word "Even"

No Comments

No Known Activations