INDEX

Explanations

references to channels in various contexts, particularly related to media and communication

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

 StatelessWidget

-0.66

ⓧ

-0.64

herzog

-0.57

 onCancelled

-0.56

catch

-0.55

는

-0.53

Lähteet

-0.53

 fact

-0.51

 buono

-0.50

afa

-0.50

POSITIVE LOGITS

 channels

1.21

Channels

1.18

 Channels

1.16

channels

1.04

 channel

0.98

 Channel

0.96

 CHANNEL

0.94

CHANNEL

0.92

 channe

0.89

 chanel

0.88

Activations Density 0.143%

references to channels in various contexts, particularly related to media and communication

No Comments

No Known Activations