INDEX

Explanations

descriptive language related to performance and quality attributes

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

describing qualities and characteristics

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

httphttps

-0.82

 snippetHide

-0.57

もしろ

-0.49

+#+#

-0.49

 betweenstory

-0.48

extAlignment

-0.45

ContentAlignment

-0.45

 createState

-0.41

thâu

-0.41

はじめに

-0.41

POSITIVE LOGITS

WithMany

0.46

AnchorStyles

0.45

ниципа

0.45

 auroit

0.43

IntoConstraints

0.43

 gröss

0.43

 zust

0.42

 ujednoznacz

0.41

 prefeitura

0.41

 avoient

0.40

Activations Density 0.033%

descriptive language related to performance and quality attributes

describing qualities and characteristics

No Comments

No Known Activations