INDEX

Explanations

actions related to participation, confrontation, and social interaction

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

actions followed by context

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

ientras

-0.58

ſelben

-0.55

 nahil

-0.55

 fashiola

-0.54

SequentialGroup

-0.54

 imagui

-0.53

 beſch

-0.53

ロウィン

-0.52

$_-

-0.52

ſſung

-0.52

POSITIVE LOGITS

AndEndTag

0.66

 HasFactory

0.40

enumi

0.38

RenderAtEndOf

0.37

antMatchers

0.37

 stanovnika

0.34

 purposes

0.33

}{@

0.33

 الحره

0.32

GraphicsUnit

0.31

Activations Density 0.053%

actions related to participation, confrontation, and social interaction

actions followed by context

No Comments

No Known Activations