INDEX

Explanations

references to specific entities or subjects within a context

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

following numbers

np_acts-logits-general · gemini-2.0-flash

Pettersson chair requested

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

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 <=",

-0.43

 Савезне

-0.40

hematical

-0.38

oredCriteria

-0.38

 chwil

-0.35

iffen

-0.35

onnaissance

-0.35

Cyfarwyddwr

-0.35

deelte

-0.34

 iyong

-0.33

POSITIVE LOGITS

 typelib

0.64

AccessorTable

0.55

 disambiguazione

0.51

 يتيمه

0.50

MemoryWarning

0.49

OOTDTY

0.49

UrlResolution

0.48

toMatchSnapshot

0.47

 Sigue

0.47

principalTable

0.46

Activations Density 0.403%

references to specific entities or subjects within a context

following numbers

Pettersson chair requested

No Comments

No Known Activations