INDEX

Explanations

structures or elements indicated by brackets and notations commonly used in research papers or academic 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

UrlResolution

-0.74

 yourself

-0.68

 sendiri

-0.67

MLLoader

-0.61

 ourselves

-0.58

InputTagHelper

-0.57

 himself

-0.57

yourself

-0.57

Yourself

-0.56

 secretos

-0.56

POSITIVE LOGITS

 footnote

0.63

 citation

0.62

note

0.60

 Heav

0.60

intios

0.59

$_"

0.58

 Mixture

0.58

citation

0.57

Note

0.57

 wherefore

0.56

Activations Density 0.158%

structures or elements indicated by brackets and notations commonly used in research papers or academic contexts

No Comments

No Known Activations