INDEX

Explanations

data related to durations and attributes of items, particularly in a structured or tabular format

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

statistical variables and foreign words

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

RectangleBorder

-0.49

kijken

-0.46

 ediciones

-0.45

 mediodía

-0.45

樣

-0.45

vención

-0.44

プライズ

-0.43



-0.43

دانشنامهٔ

-0.43

attamento

-0.43

POSITIVE LOGITS

AddHtmlAttribute

0.53

adaptiveStyles

0.48

جغرافيا

0.42

windowFixed

0.42

ніципалі

0.40

kháu

0.40

RTEX

0.40

 censiti

0.39

Spoljašnje

0.39

expandindo

0.38

Activations Density 0.037%

data related to durations and attributes of items, particularly in a structured or tabular format

statistical variables and foreign words

No Comments

No Known Activations

data related to durations and attributes of items, particularly in a structured or tabular format

statistical variables and foreign words

No Comments

No Known Activations