INDEX

Explanations

references to specific TV shows and characters, particularly focusing on themes related to medical dramas

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

TV shows and procedural dramas

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

rawdę

-0.53

itoare

-0.49

 îng

-0.48

 հղումներ

-0.47

 Versammlung

-0.47

 burbujas

-0.47

 învă

-0.47

 Económica

-0.46

 hablamos

-0.46

 scattata

-0.46

POSITIVE LOGITS

 Jovi

0.46

 shift

0.45

 rotation

0.43

 Rotation

0.43

 Shift

0.42

TextWatcher

0.40

ROT

0.39

Shift

0.39

sal

0.38

 synd

0.38

Activations Density 0.017%

references to specific TV shows and characters, particularly focusing on themes related to medical dramas

TV shows and procedural dramas

No Comments

No Known Activations