INDEX

Explanations

references to the word "punch" in various contexts

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

punch (tool, drink, strike)

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

 heilig

-0.44

atoare

-0.44

 mesmas

-0.43

didSet

-0.43

 semula

-0.42

 mismas

-0.42

 liggen

-0.41

 erwartet

-0.41

 było

-0.41

 leeftijd

-0.41

POSITIVE LOGITS

 punch

1.34

 Punch

1.30

Punch

1.25

punch

1.21

 PUNCH

1.16

 punches

1.05

 punched

0.96

 punching

0.95

 Pank

0.79

punctuation

0.71

Activations Density 0.006%

references to the word "punch" in various contexts

punch (tool, drink, strike)

No Comments

No Known Activations