INDEX

Explanations

terms related to positive changes or improvements

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

that describe chance or modification (reduce, increase, improve, alter)

np_acts-logits-general · gemini-2.0-flash

change or modification

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

 EconPapers

-0.36

no

-0.31

modelBuilder

-0.31

 []).

-0.30

bau

-0.30

jewództ

-0.30

me

-0.27

see

-0.26

ref

-0.26

behörde

-0.25

POSITIVE LOGITS

GOTREF

0.79

AlterField

0.75

posedge

0.74

rungsseite

0.71

transQ

0.70

httphttps

0.68

#+#

0.68

 snippetHide

0.67

aarrggbb

0.66

 Chwiliwch

0.65

Activations Density 0.492%

terms related to positive changes or improvements

that describe chance or modification (reduce, increase, improve, alter)

change or modification

No Comments

No Known Activations