INDEX

Explanations

the presence of a document's beginning or introduction markers

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

Appears before uncommon words/tokens

np_acts-logits-general · gemini-2.0-flash

months until

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 весьма

-0.69

Perhaps

-0.68

perhaps

-0.68

 perhaps

-0.64

 jenen

-0.62

(!)

-0.61

eraard

-0.60

 Perhaps

-0.59

 jenem

-0.59

 "...

-0.59

POSITIVE LOGITS

 Cancelled

0.65

 Quora

0.65

 narcissist

0.63

Rüyada

0.62

setupUi

0.61

 GREY

0.61

StoryboardSegue

0.61

 Mongols

0.61

adomo

0.58

afficheront

0.58

Activations Density 0.291%

the presence of a document's beginning or introduction markers

Appears before uncommon words/tokens

months until

No Comments

No Known Activations