INDEX

Explanations

references to directions and specific scientific measurements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 0-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_0/width_65k/average_l0_73

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.0.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

AnimationsModule

-0.70

 snippetHide

-0.66

 raiſ

-0.65

 voli

-0.62

 Houſe

-0.61

 chofe

-0.60

 onely

-0.59

 setShow

-0.59

 Hebrews

-0.59

ighbor

-0.59

POSITIVE LOGITS

Tester

0.69

 "..\..\

0.63

ktm

0.63

 "..\..\..\

0.59

 dehydrogenase

0.57

íslu

0.54

tester

0.51

wikimedia

0.51

tisgarh

0.51

stä

0.50

Activations Density 0.111%

references to directions and specific scientific measurements

No Comments

No Known Activations