INDEX

Explanations

terms and structures related to programming constructs and data management within code

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

Code and numbers

np_acts-logits-general · gemini-2.0-flash

repeated leadership styles

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

chap

-0.54

guer

-0.49

 "..\..\

-0.48

pezifische

-0.47

üren

-0.46

vol

-0.45

 <<<<<<<<<<<<<<

-0.45

mbres

-0.45

 shout

-0.44

POSITIVE LOGITS

IsContent

0.76

 absl

0.66



0.64

 jsPsych

0.60

 gyhoeddwyd

0.60

 ligiloj

0.58

 espagne

0.57

ReusableCell

0.56

ForTesting

0.56

 població

0.56

Activations Density 0.406%

terms and structures related to programming constructs and data management within code

Code and numbers

repeated leadership styles

No Comments

No Known Activations