INDEX

Explanations

phrases related to making requests or asking for something

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 11-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_11/width_16k/average_l0_80

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.11.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

versaries

-0.58

TypeDef

-0.57

/>";

-0.56

 Crick

-0.54

 Pence

-0.54

ulet

-0.52

HOUT

-0.52

 latch

-0.50

upar

-0.49

дента

-0.49

POSITIVE LOGITS

 requests

1.27

 Requests

1.18

 request

1.17

 requested

1.16

 requesting

1.16

 Request

1.12

 REQUEST

1.10

REQUEST

1.07

 demands

1.00

requests

0.98

Activations Density 0.260%

phrases related to making requests or asking for something

No Comments

No Known Activations