INDEX

Explanations

specific terms and phrases related to the dynamics of a racing event

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

race course segments and laps

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

 Normdatei

-0.90

IUrlHelper

-0.88

ModelAdmin

-0.84

 ſind

-0.82

OGND

-0.82

<unused41>

-0.77

<unused42>

-0.77

<unused8>

-0.77

[@BOS@]

-0.77

<unused3>

-0.77

POSITIVE LOGITS

 approaching

0.30

run

0.27

ToRefresh

0.26

 tắt

0.26

wehr

0.24

 nearing

0.24

 veril

0.23

 ramp

0.23

 пом

0.23

 pump

0.22

Activations Density 0.006%

specific terms and phrases related to the dynamics of a racing event

race course segments and laps

No Comments

No Known Activations