INDEX

Explanations

references to figures and tables in the document

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

fig references

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

begin

-0.41

 franchise

-0.34

WriteTagHelper

-0.33

ленность

-0.33

delwed

-0.33

liferay

-0.32

 الحره

-0.32

 Franch

-0.31

 незавершена

-0.31

исленность

-0.31

POSITIVE LOGITS

graphicx

0.65

setVerticalGroup

0.61

Figure

0.61

 fotografico

0.61

 Figures

0.59

 FIGURES

0.58

Figures

0.57

IntoConstraints

0.56

 ModelExpression

0.56

 Figs

0.55

Activations Density 0.597%

references to figures and tables in the document

fig references

No Comments

No Known Activations