INDEX

Explanations

references to shirts

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

shirt

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ronpa

-0.53

allan

-0.52

Max

-0.51

 Grand

-0.50

 numerus

-0.49

 Moreau

-0.49

OGA

-0.48

GV

-0.48

BV

-0.48

 Gregorio

-0.48

POSITIVE LOGITS

 Shirt

1.21

Shirt

1.16

 shirt

1.15

Shirts

1.04

shirt

1.02

 shirts

1.00

 Shirts

0.98

shirts

0.89

hirt

0.85

 Monfieur

0.85

Activations Density 0.004%

references to shirts

shirt

No Comments

No Known Activations