INDEX

Explanations

formatted time or timestamp information

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

">//

-0.89

 itſelf

-0.76

GTO

-0.68

 Bourgoin

-0.66

 LXXX

-0.64

 oaths

-0.63

 rattan

-0.63

 houſe

-0.63

UAV

-0.60

ICF

-0.60

POSITIVE LOGITS

 ActionBar

0.51

enumi

0.51

EndProject

0.51

لينكات

0.51

DoubleQuotes

0.50

onAttach

0.49

 once

0.49

enas

0.49

edoc

0.48

Activations Density 0.599%

formatted time or timestamp information

No Comments

No Known Activations