INDEX

Explanations

references to transformations and related processes

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-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

eyed

-0.48

 Gries

-0.47

ey

-0.45

 getID

-0.44

ee

-0.44

eye

-0.44

QQ

-0.43

 Breen

-0.42

ii

-0.42

eyes

-0.41

POSITIVE LOGITS

transform

1.02

Transform

1.00

 Transform

0.98

TRANSFORM

0.85

 transform

0.84

 MonoBehaviour

0.84

 TRANSFORM

0.77

setTransform

0.76

transforms

0.72

 transforms

0.70

Activations Density 0.017%

references to transformations and related processes

No Comments

No Known Activations