INDEX

Explanations

questions about user actions or requirements regarding specific fixes and features in a system

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

Followed by "is"

np_acts-logits-general · gemini-2.0-flash

sentence endings

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

ſelf

-0.75

neſs

-0.72

^(@)

-0.71

 chofe

-0.66

 Jefus

-0.65

 fhew

-0.64

 auffi

-0.63

 myſelf

-0.63

 pleaſure

-0.63

Ꮎ

-0.62

POSITIVE LOGITS

 propOrder

0.84

<eos>

0.82

 Audiodateien

0.75

 EconPapers

0.73

RenderAtEndOf

0.72

…

0.71

richTextPanel

0.71

rungsseite

0.71

Билгалдахарш

0.71

)|^{

0.70

Activations Density 0.141%

questions about user actions or requirements regarding specific fixes and features in a system

Followed by "is"

sentence endings

No Comments

No Known Activations