INDEX

Explanations

phrases related to the timing of events or actions

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

time of writing/study/visit

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

ITUD

-0.41

 SPDX

-0.40

Дереккөздер

-0.39

########.

-0.37

 AspNetCore

-0.36

Rep

-0.36

 Rakyat

-0.35

RegressionTest

-0.35

 hjälp

-0.35

 كورة

-0.35

POSITIVE LOGITS

 للاسماء

0.55

 Chwiliwch

0.50

localctx

0.49

ьаж

0.49

InitVars

0.47

 ویکی‌پدی

0.46

principalTable

0.46

consulté

0.46

UrlResolution

0.44

EDEFAULT

0.43

Activations Density 0.018%

phrases related to the timing of events or actions

time of writing/study/visit

No Comments

No Known Activations