INDEX

Explanations

calls to action, particularly related to reading and checking out content

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

ImageContext

-0.86

 photolibrary

-0.80

 <=",

-0.79

JpaRepository

-0.78

ngdoc

-0.75

 Shakspeare

-0.75

 raiſ

-0.74

 ―――――

-0.74

UserScript

-0.73

 cdti

-0.73

POSITIVE LOGITS

 full

0.95

 more

0.86

all

0.79

our

0.78

 latest

0.72

 peł

0.69

my

0.68

full

0.67

 complete

0.66

 Full

0.65

Activations Density 0.145%

calls to action, particularly related to reading and checking out content

No Comments

No Known Activations