INDEX

Explanations

references to legal issues and consequences

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

after specific abbreviations or symbols

np_acts-logits-general · gemini-2.0-flash

type, city, root, possessive pronouns

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

-0.73

-0.70

“

-0.68

<eos>

-0.68

‘

-0.67

…

-0.65

-0.62

-0.61

...

-0.60

in

-0.59

POSITIVE LOGITS

 myſelf

1.54

 itſelf

1.39

 pleaſure

1.35

 auffi

1.33

 purpoſe

1.33

 Jefus

1.32

 themſelves

1.32

 raiſ

1.28

 ſche

1.28

 himſelf

1.27

Activations Density 0.058%

references to legal issues and consequences

after specific abbreviations or symbols

type, city, root, possessive pronouns

No Comments

No Known Activations