INDEX

Explanations

references to "frame" 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

 Majefty

-0.78

$$\

-0.70

 aéreas

-0.67

 Okada

-0.67

DispatchToProps

-0.67

^(@)

-0.66

$_"

-0.66

سكري

-0.66

первых

-0.65

 Alcott

-0.65

POSITIVE LOGITS

 frames

1.03

 Frames

0.99

frames

0.95

Frames

0.89

 Frame

0.78

works

0.76

 FRAME

0.75

less

0.74

 frame

0.73

 フレーム

0.72

Activations Density 0.094%

references to "frame" in various contexts

No Comments

No Known Activations