INDEX

Explanations

numerical values and references to academic articles or citations

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

Commas followed by single letters

np_acts-logits-general · gemini-2.0-flash

names with initials

np_acts-logits-general · gemini-2.5-flash-lite

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ArgumentParser

-0.53

AxisAlignment

-0.53

raisemb

-0.51

gut

-0.47

 Flax

-0.47

 Spir

-0.47

Geplaatst

-0.47

ULT

-0.47

GUT

-0.46

Gav

-0.46

POSITIVE LOGITS

 الحره

0.72

IUrlHelper

0.63

SqlClient

0.60

 OBITUARY

0.60

0.59

ETHING

0.59

abetes

0.58

 JAXBElement

0.58

Activations Density 0.237%

numerical values and references to academic articles or citations

Commas followed by single letters

names with initials

No Comments

No Known Activations