INDEX

Explanations

structured data representations and their properties

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

 кӀош

-0.59

 representan

-0.47

นะครับ

-0.47

srand

-0.46

่น

-0.45

kampf

-0.44

}\|

-0.44

Tämä

-0.44

Paged

-0.43

kyl

-0.43

POSITIVE LOGITS

featureID

0.91

tagHelperRunner

0.84

uxxxx

0.83



0.82

Personendaten

0.81

httphttps

0.78

unmodifiable

0.75

ьаж

0.74

IContainer

0.73

 pinulongan

0.73

Activations Density 0.601%

structured data representations and their properties

No Comments

No Known Activations