INDEX

Explanations

characters and symbols from various languages or scripts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

 Egipto

-0.56

 knji

-0.56

 mármol

-0.53

 disfraces

-0.49

 práctico

-0.48

 aliento

-0.48

 keç

-0.48

tvguidetime

-0.47

 keben

-0.47

 maravillas

-0.47

POSITIVE LOGITS

 لينك

0.63

 للمعارف

0.60

mybatisplus

0.57

parsedMessage

0.57

TryDecodeAsNil

0.54

 ſte

0.52

 autob

0.52

 ſur

0.51

 مشين

0.50

 faſt

0.50

Activations Density 0.044%

characters and symbols from various languages or scripts

No Comments

No Known Activations