INDEX

Explanations

references to deer and their various types or characteristics

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

deer

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

Format

-0.57

 tắc

-0.50

Massimo

-0.50

AMT

-0.50

 Whitfield

-0.48

Frank

-0.47

 Maxi

-0.47

CSP

-0.47

 Barth

-0.47

Sonia

-0.47

POSITIVE LOGITS

 Deer

1.17

Deer

1.13

 deer

1.09

deer

0.89

 venison

0.67

🦌

0.63

鹿

0.63

indeer

0.62

 cerv

0.61

鹿

0.60

Activations Density 0.004%

references to deer and their various types or characteristics

deer

No Comments

No Known Activations