INDEX

Explanations

statements related to code inclusion or library imports

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

 vPvB

-0.70

vœ

-0.65

 Infór

-0.60

transQ

-0.60

 ſch

-0.60

cæ

-0.59

 MatDialog

-0.58

 nôtre

-0.56

 vôtre

-0.55

 consommate

-0.54

POSITIVE LOGITS

include

1.49

import

0.86

Include

0.80

 include

0.72

 Include

0.69

INCLUDE

0.66

includes

0.66

 includes

0.60

Includes

0.60

AnchorStyles

0.58

Activations Density 0.019%

statements related to code inclusion or library imports

No Comments

No Known Activations