INDEX

Explanations

phrases indicating suitability or ideal conditions for various activities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 23-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_23/width_16k/average_l0_120

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.23.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

 bijzondere

-0.47

 spéciaux

-0.46

 werkzaam

-0.45

 strijd

-0.44

 särsk

-0.43

נטרנט

-0.42

 zieken

-0.41

\{\\

-0.41

 goederen

-0.40

 noqa

-0.40

POSITIVE LOGITS

 suited

0.60

suited

0.52

 dandy

0.51

SequentialGroup

0.50

Nice

0.48

 Nice

0.47

 ideal

0.47

mix

0.47

 intptr

0.47

 Ideal

0.47

Activations Density 0.376%

phrases indicating suitability or ideal conditions for various activities

No Comments

No Known Activations