INDEX

Explanations

information about organizational entities, including events and affiliations

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

Text after commas or parentheses

np_acts-logits-general · gemini-2.0-flash

which,

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

tonode

-0.62

 ddelweddau

-0.55

 umgekehrt

-0.52

utnik

-0.51

ắp

-0.51

vailability

-0.51

 Réalisation

-0.49

oughby

-0.49

theless

-0.48

 priori

-0.47

POSITIVE LOGITS

 which

1.32

which

1.09

formerly

1.00

 formerly

0.99

itself

0.83

 WHICH

0.80

aka

0.80

Which

0.78

 whose

0.77

 sebuah

0.77

Activations Density 0.506%

information about organizational entities, including events and affiliations

Text after commas or parentheses

which,

No Comments

No Known Activations