INDEX

Explanations

phrases related to identification and processing in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 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

nameof

-0.47

 enfans

-0.44

part

-0.41

__(

-0.40

客様

-0.37

 EClass

-0.34

CPS

-0.34

gebn

-0.33

instancetype

-0.33

it

-0.33

POSITIVE LOGITS

 various

0.68

 különböző

0.62

 pinulongan

0.58

 forskellige

0.57

 numerous

0.57

 różne

0.57

 berbagai

0.57

各种

0.56

 Wikimedijinoj

0.56

各类

0.56

Activations Density 0.726%

phrases related to identification and processing in various contexts

No Comments

No Known Activations