INDEX

Explanations

references to specific health conditions or treatments

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

ending in "el" or "ol"

np_acts-logits-general · gemini-2.0-flash

tokens ending in `el` or `vel`

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 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

ed

-1.23

으로

-1.03

-0.88

ا

-0.85

ians

-0.77

ي

-0.72

ih

-0.69

ian

-0.69

ie

-0.67

POSITIVE LOGITS

led

0.84

Zol

0.77

 CURIAM

0.77

ligence

0.73

GEBURTSDATUM

0.70

Lel

0.70

таратура

0.69

ppure

0.69

 يتيمه

0.69

ล์

0.69

Activations Density 0.620%

references to specific health conditions or treatments

ending in "el" or "ol"

tokens ending in `el` or `vel`

No Comments

No Known Activations