INDEX

Explanations

programming annotations and code structures related to functions

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

Code-related terms and function calls

np_acts-logits-general · gemini-2.0-flash

try followed by code or command

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

tigma

-0.66

Климат

-0.60

NewGuid

-0.60

 GBIF

-0.53

<eos>

-0.48

hnia

-0.47

 кӀ

-0.47

wikk

-0.47

käse

-0.46

Panamoan

-0.46

POSITIVE LOGITS

 referrerpolicy

0.59

 EconPapers

0.57

 Normdatei

0.57

IsContent

0.56

Tembelea

0.54

 nakalista

0.54

kurat

0.53

TestBed

0.53

AxisAlignment

0.53

(;;)

0.51

Activations Density 0.871%

programming annotations and code structures related to functions

Code-related terms and function calls

try followed by code or command

No Comments

No Known Activations