INDEX

Explanations

terms related to changes or modifications

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

alter

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

 TextAppearance

-0.62

busch

-0.61

AndroidJUnit

-0.61

 sessionId

-0.60

 soybeans

-0.60

 مرئيه

-0.59

mybatisplus

-0.58

ManagedBean

-0.58

 Economies

-0.57

────────

-0.57

POSITIVE LOGITS

 alteration

1.52

 alterations

1.42

 altered

1.41

 Alter

1.41

 alter

1.39

altered

1.37

Alter

1.36

 altering

1.30

alter

1.29

 Altered

1.21

Activations Density 0.016%

terms related to changes or modifications

alter

No Comments

No Known Activations