INDEX

Explanations

terms related to electrical components and circuits

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

figures (e.g., FIG. 1)

np_acts-logits-general · gemini-2.0-flash

FIG. and numbers

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

InstrumentedTest

-0.64

sizeCache

-0.57

+:+

-0.56

 NSCoder

-0.49

 sociali

-0.45

toBeTruthy

-0.45

 means

-0.45

 veau

-0.44

 consultato

-0.44

httphttps

-0.44

POSITIVE LOGITS

 препратки

0.74

 uParam

0.69

saraba

0.63

μως

0.63

 Poste

0.62

pidou

0.61

izr

0.60

 Préférences

0.60

IContainer

0.58

 ***!

0.57

Activations Density 0.531%

terms related to electrical components and circuits

figures (e.g., FIG. 1)

FIG. and numbers

No Comments

No Known Activations