INDEX

Explanations

references to articles or documentation

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

user article/page

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-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

цездатний

-0.47

aux

-0.43

get

-0.43

to

-0.42

alt

-0.41

save

-0.41

-0.40

store

-0.40

destroyed

-0.39

POSITIVE LOGITS

CloseOperation

0.70

MemoryWarning

0.65

 Infórmanos

0.63

 Signalez

0.60

ロウィン

0.57

 queſta

0.57

rungsseite

0.57

 חיצוניים

0.56

 ContentValues

0.55

tagHelperRunner

0.54

Activations Density 0.000%

references to articles or documentation

user article/page

No Comments

No Known Activations