INDEX

Explanations

occurrences of phrases and items referenced in a sequential or singular context

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

Followed by "a" or a comma

np_acts-logits-general · gemini-2.0-flash

each unit

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

itteet

-0.52

moire

-0.49

INNER

-0.49

Aspect

-0.48

自身

-0.45

 Fakult

-0.45

soby

-0.45

 جز

-0.44

 juga

-0.43

inner

-0.43

POSITIVE LOGITS

 apiece

1.47

씩

1.05

 each

1.03

ずつ

0.98

 ciascuno

0.94

 saja

0.93

PYX

0.91

each

0.91

 виправивши

0.88

 ciasc

0.87

Activations Density 0.432%

occurrences of phrases and items referenced in a sequential or singular context

Followed by "a" or a comma

each unit

No Comments

No Known Activations