INDEX

Explanations

words related to prevention

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

preventing actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_31/width_131k/average_l0_109

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.31.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

 guineas

-0.42

 referenties

-0.39

argout

-0.38

 vermelhas

-0.38

usermodel

-0.37

мәләр

-0.35

iciary

-0.35

хьтан

-0.35

 marchés

-0.34

POSITIVE LOGITS

Prevent

0.73

prevent

0.69

 Prevent

0.65

 Prevents

0.65

 prevent

0.65

 autorytatywna

0.64

 <<<<<<<<<<<<<<

0.64

 terjadinya

0.61

 prevents

0.61

 PREVENT

0.60

Activations Density 0.018%

words related to prevention

preventing actions

No Comments

No Known Activations