INDEX

Explanations

mathematical symbols and expressions related to equations and functions

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

bend right/left

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

ChildScrollView

-0.86

 houſe

-0.81

 Houſe

-0.69

numerusform

-0.69

bootstrapcdn

-0.68

WriteTagHelper

-0.66

 ſmall

-0.66

 pleaſure

-0.66

 purpoſe

-0.66

 Reſ

-0.66

POSITIVE LOGITS

 nakalista

0.34

 causar

0.31

 enamorado

0.30

 humild

0.29

 robar

0.29

tagHelperRunner

0.29

 рассе

0.29

 silicona

0.28

 olhando

0.28

 coger

0.28

Activations Density 0.209%

mathematical symbols and expressions related to equations and functions

bend right/left

No Comments

No Known Activations