INDEX

Explanations

temporal or sequential relationships in text

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

positions of words

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

inSlope

-0.38

chism

-0.36

 houſe

-0.34

.*")]

-0.33

 bagay

-0.32

 jdbcTemplate

-0.32

izacin

-0.32

 dedans

-0.32

PendingIntent

-0.31

msgTypes

-0.31

POSITIVE LOGITS

 مباشرة

0.60

uxxxx

0.58

 հղումներ

0.57

 ब्रेकडाउन

0.55

webElementXpaths

0.54

 eigentlichen

0.54

 INTERESAR

0.54

verifyException

0.50

WriteLiteral

0.50

exitRule

0.49

Activations Density 0.073%

temporal or sequential relationships in text

positions of words

No Comments

No Known Activations