INDEX

Explanations

references to specific entities and clarify their context or relationships within the text

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

antibody, observed, presented, stories

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

 snippetHide

-0.51

RenderAtEndOf

-0.45

 informée

-0.44

 Мексичка

-0.43

JpaRepository

-0.43

zelfde

-0.42

 âmes

-0.42

AxisAlignment

-0.42

 enfans

-0.42

WriteTagHelper

-0.42

POSITIVE LOGITS

unków

0.38

 useParams

0.37

épendance

0.36

 uska

0.35

rhein

0.35

HasColumnName

0.35

 whose

0.35

Ab

0.33

 столь

0.33

Activations Density 0.132%

references to specific entities and clarify their context or relationships within the text

antibody, observed, presented, stories

No Comments

No Known Activations