INDEX

Explanations

code-related syntax and definitions

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

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Tuttle

-0.63

 مشين

-0.58

 defStyleAttr

-0.58

 hjer

-0.56

 parachute

-0.54

appartamento

-0.54

 endforeach

-0.54

entliche

-0.54

 Fernandez

-0.53

 abstracta

-0.53

POSITIVE LOGITS

ing

0.84

ding

0.84

lling

0.82

mapping

0.80

eling

0.80

ning

0.79

ADING

0.78

pting

0.78

oping

0.77

 batting

0.76

Activations Density 0.302%

code-related syntax and definitions

No Comments

No Known Activations