INDEX

Explanations

questions or phrases querying information

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 4-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_4/width_16k/average_l0_124

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.4.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

 IndexError

-0.70

 Avalon

-0.69

 Cyclops

-0.69

 Balik

-0.69

สือ

-0.68

 Bamberg

-0.67

 KeyError

-0.66

 Moors

-0.65

 Blok

-0.65

 Brea

-0.65

POSITIVE LOGITS

 what

1.45

what

1.19

 WHAT

1.13

 What

1.05

WHAT

1.03

What

0.99

 hvad

0.77

wat

0.69

ArgsConstructor

0.69

hva

0.68

Activations Density 0.150%

questions or phrases querying information

No Comments

No Known Activations