INDEX

Explanations

numbers and alphanumeric codes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 beauté

-0.53

"}")

-0.51

 faccia

-0.50

 bellezza

-0.49

sendRedirect

-0.49

"}"

-0.49

 récompenses

-0.48

sikan

-0.47

нительно

-0.46

 belleza

-0.46

POSITIVE LOGITS

 betweenstory

0.62

NUMX

0.61

⿷

0.61

AntiForgeryToken

0.59

 gratu

0.59

 pail

0.59

 SwitchCompat

0.59

NewLabel

0.58

 Partici

0.57

 CURIAM

0.56

Activations Density 0.303%

numbers and alphanumeric codes

No Comments

No Known Activations