INDEX

Explanations

references to body parts and associated actions

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

Tokens preceding "head" or "heads"

np_acts-logits-general · gemini-2.0-flash

body parts ending in head

np_acts-logits-general · gemini-2.5-flash-lite

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Kariera

-0.56

цездатний

-0.50

 lenker

-0.49

 Bacteriol

-0.49

loroethene

-0.48

 Fong

-0.47

XSS

-0.47

tisseur

-0.47

Abitanti

-0.46

prüche

-0.46

POSITIVE LOGITS

 head

2.71

 heads

2.46

 Head

2.36

head

2.32

Head

2.32

 HEAD

2.24

HEAD

2.10

 Heads

2.09

heads

2.08

Heads

1.98

Activations Density 0.291%

references to body parts and associated actions

Tokens preceding "head" or "heads"

body parts ending in head

No Comments

No Known Activations