INDEX

Explanations

references to URLs and web resources

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

adpleegd

-0.66

expandindo

-0.63

KURZBESCHREIBUNG

-0.58

ConstraintMaker

-0.54

 AssemblyTitle

-0.54

 resourceCulture

-0.53

LikeLiked

-0.52

baomidou

-0.51

unknownFields

-0.51

Pozri

-0.51

POSITIVE LOGITS

index

0.96

Default

0.78

 index

0.78

Detail

0.77

default

0.76

Index

0.75

cgi

0.72

 Archived

0.72

servlet

0.72

 Theſe

0.71

Activations Density 0.274%

references to URLs and web resources

No Comments

No Known Activations