INDEX

Explanations

instances of various articles and descriptors preceding nouns, indicating a focus on entities and their classifications

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

following the indefinite article "a" or "an"

np_acts-logits-general · gemini-2.0-flash

determiner followed by noun

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 28-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_28/width_16k/average_l0_119

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.28.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

 emperador

-0.35

The

-0.31

<bos>

-0.31

 Following

-0.29

Fazit

-0.28

 yeterli

-0.28

 reempla

-0.26

 true

-0.26

 abraz

-0.26

 muka

-0.25

POSITIVE LOGITS

ロウィン

0.91

 EconPapers

0.86

EDEFAULT

0.85

 '\\;'

0.81

<unused23>

0.79

[@BOS@]

0.79

<unused43>

0.79

niſſe

0.79

<unused1>

0.79

<unused3>

0.79

Activations Density 0.585%

instances of various articles and descriptors preceding nouns, indicating a focus on entities and their classifications

following the indefinite article "a" or "an"

determiner followed by noun

No Comments

No Known Activations

instances of various articles and descriptors preceding nouns, indicating a focus on entities and their classifications

following the indefinite article "a" or "an"

determiner followed by noun

No Comments

No Known Activations