INDEX

Explanations

technical or mathematical terminology

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 1-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_1/width_16k/average_l0_102

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.1.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

useEffect

-0.65

 faits

-0.63

 للاسماء

-0.61

FYI

-0.57

منی

-0.55

 obra

-0.55

FYI

-0.54

Serial

-0.54

 chapa

-0.54

Trait

-0.52

POSITIVE LOGITS

 annual

0.80

Gambas

0.71

Linq

0.69

htë

0.69

şekkür

0.68

IntoConstraints

0.66

annual

0.65

 minimum

0.64

betical

0.64

 jähr

0.61

Activations Density 0.086%

technical or mathematical terminology

No Comments

No Known Activations