INDEX

Explanations

positive adjectives and expressions of approval

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

positive adjectives

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

TypedDataSet

-0.49

EDEFAULT

-0.48

RTLR

-0.45

RenderAtEndOf

-0.44

下载附件

-0.43

 estekak

-0.43

 informée

-0.43

 Himo

-0.42

 NSCoder

-0.42

%)$

-0.41

POSITIVE LOGITS

 awesome

0.77

 cool

0.73

Cool

0.68

 Awesome

0.68

 Cool

0.67

 AWESOME

0.65

Awesome

0.65

 coolest

0.64

cool

0.63

awesome

0.60

Activations Density 0.027%

positive adjectives and expressions of approval

positive adjectives

No Comments

No Known Activations