INDEX

Explanations

words and phrases related to unexpected success or achievements

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

remarkable events and outcomes

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

 this

-0.30

ạy

-0.30

 souverain

-0.30

lze

-0.30

 løpet

-0.29

 prossimo

-0.29

 benne

-0.29

Too

-0.29

 prí

-0.29

POSITIVE LOGITS

IsContent

0.77

CppMethod

0.74

:✨

0.70

sizeCache

0.68

tvguidetime

0.68

contentLoaded

0.66

HostException

0.65

 الحره

0.65

 otomatig

0.65

httphttps

0.63

Activations Density 0.028%

words and phrases related to unexpected success or achievements

remarkable events and outcomes

No Comments

No Known Activations