INDEX

Explanations

significant descriptors or adjectives that characterize a situation or context

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

citations and special characters

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

-0.38

is

-0.35

 next

-0.31

 округа

-0.29

 delivery

-0.29

 cause

-0.28

 itself

-0.28

 Lenz

-0.28

 tarifs

-0.28

 feature

-0.27

POSITIVE LOGITS

 OMITBAD

0.79

 ſondern

0.77

ロウィン

0.77

 ſehen

0.77

wiliwch

0.77

 يتيمه

0.76

majánló

0.76

niſſe

0.76

 Menſchen

0.75

 ſind

0.75

Activations Density 0.044%

significant descriptors or adjectives that characterize a situation or context

citations and special characters

No Comments

No Known Activations