INDEX

Explanations

references to vague or unspecified concepts or entities

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

something followed by continuations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 16-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_16/width_65k/average_l0_128

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.16.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

Datuak

-0.88

UnusedPrivate

-0.69

AutoScaleMode

-0.69

 Arquivado

-0.68

lye

-0.68

 Kays

-0.67

xase

-0.67

prefixer

-0.65

Aktualisiert

-0.64

 foncé

-0.63

POSITIVE LOGITS

ETHING

1.17

 else

1.12

 something

1.09

Something

1.08

 Something

1.07

Somebody

1.05

something

1.05

 SOMETHING

1.03

Someone

0.93

 somethin

0.85

Activations Density 0.030%

references to vague or unspecified concepts or entities

something followed by continuations

No Comments

No Known Activations