INDEX

Explanations

code-related terms or keywords

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

-0.40

ans

-0.39

<table>

-0.38

-0.36

ons

-0.36

ont

-0.35

 molta

-0.35

iz

-0.34

kur

-0.33

POSITIVE LOGITS

 gainera

0.91

AndEndTag

0.90

 autorytatywna

0.86

prefixer

0.85

 propOrder

0.84

findpost

0.84

enumi

0.84

 esternos

0.82

UserScript

0.82

UnusedPrivate

0.81

Activations Density 0.067%

code-related terms or keywords

No Comments

No Known Activations