INDEX

Explanations

phrases related to official statements and communications

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 10-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_10/width_16k/average_l0_77

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.10.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

kháu

-0.63

 zvuky

-0.62

 corris

-0.57

 nemico

-0.56

 insegna

-0.55

 desiderio

-0.55

 Cientí

-0.55

 gradova

-0.54

 käytet

-0.54

 landings

-0.54

POSITIVE LOGITS

脚注の使い方

0.67

 interview

0.64

SequentialGroup

0.59

 told

0.58

 интервью

0.56

OCI

0.55

 interviewed

0.55

 gegenüber

0.55

IActionResult

0.54

 eigener

0.53

Activations Density 0.142%

phrases related to official statements and communications

No Comments

No Known Activations