INDEX

Explanations

terminology related to scaling in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 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

🏻

-0.65

—

-0.57

ednesday

-0.57

__*/

-0.54

凰

-0.52

 propOrder

-0.52

ly

-0.52

spricht

-0.52

)");

-0.51

CompilerServices

-0.51

POSITIVE LOGITS

 Scales

1.04

 scales

1.02

 SCALE

0.98

 Scal

0.97

Scales

0.92

scales

0.87

 Scale

0.83

 Skal

0.79

SCALE

0.78

Scale

0.78

Activations Density 0.055%

terminology related to scaling in various contexts

No Comments

No Known Activations