INDEX

Explanations

the word "like" used in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 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

 CURIAM

-0.52

 beleid

-0.52

 ownership

-0.47

 gezet

-0.46

rogels

-0.46

//

-0.45

 Ownership

-0.45

 endeavours

-0.44

Enquiries

-0.42

 volontà

-0.42

POSITIVE LOGITS

 LIKE

1.21

Like

1.20

like

1.20

 Like

1.13

 like

1.09

LIKE

1.05

likes

0.77

 seperti

0.75

comme

0.73

Comme

0.73

Activations Density 0.147%

the word "like" used in various contexts

No Comments

No Known Activations