INDEX

Explanations

instances of the word "ко" and its occurrences in various contexts related to detection or measurement

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

ко, co, deco

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_31/width_131k/average_l0_109

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.31.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

Hiện

-0.40

 Signalez

-0.40

 Aufenthalts

-0.38

 Wilayah

-0.38

 Miß

-0.38

 sierpnia

-0.36

 botellas

-0.36

 actualidad

-0.36

 hadas

-0.36

 vuel

-0.36

POSITIVE LOGITS

 &___

0.59

 ErrIntOverflow

0.54

AddTagHelper

0.54

oneofs

0.51

co

0.51

FieldDescriptor

0.50

 coch

0.49

 scratched

0.48

 setCode

0.48

providedIn

0.47

Activations Density 0.005%

instances of the word "ко" and its occurrences in various contexts related to detection or measurement

ко, co, deco

No Comments

No Known Activations