INDEX

Explanations

segments of structured data or code, particularly those with a consistent format or syntactic structure

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

C code diff additions

np_acts-logits-general · gemini-2.0-flash

code delimiters

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-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

 considérons

-0.48

ihnachts

-0.46

 AllAfrica

-0.46

windowFixed

-0.45

Identyfik

-0.44

 regalías

-0.44

shadowOpacity

-0.44

tvguidetime

-0.43

 představ

-0.43

 Kowal

-0.43

POSITIVE LOGITS

[toxicity=0]

0.87

__(/*!

0.77

SBATCH

0.57

⦑

0.57

/******/

0.55

ActionEvent

0.55

 pylint

0.54

 }^{*}$

0.54

⁩

0.54

 Gabor

0.53

Activations Density 0.333%

segments of structured data or code, particularly those with a consistent format or syntactic structure

C code diff additions

code delimiters

No Comments

No Known Activations