INDEX

Explanations

significant actions or events that are often described with strong punctuation or context

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

phrases ending a sentence, listing or description

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

 Sampson

-0.44

バリ

-0.40

 baffled

-0.37

 zbyt

-0.36

 Samson

-0.36

sam

-0.36

 sceptical

-0.36

inf

-0.36

odzi

-0.36

öffel

-0.35

POSITIVE LOGITS

 disambiguazione

0.71

CppMethod

0.68

rungsseite

0.62

 <<<<<<<<<<<<<<

0.56

#+#

0.54

 ComVisible

0.52

RuleContext

0.51

 يتيمه

0.51

 ujednoznacz

0.49

protoimpl

0.48

Activations Density 0.002%

significant actions or events that are often described with strong punctuation or context

phrases ending a sentence, listing or description

No Comments

No Known Activations