INDEX

Explanations

punctuation marks and sentence endings

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

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

 muß

-0.86

・・・・・

-0.85

 läßt

-0.82

・・・・

-0.81

 müßte

-0.69



-0.66

ㄋ

-0.65

 .......

-0.65

 unsurpassed

-0.65

 mußte

-0.64

POSITIVE LOGITS

Idk

1.08

 shitty

1.05

idk

1.05

 fucked

1.03

 fucking

1.03

idk

1.02

tbh

1.02

 goddamn

0.99

 FUCKING

0.98

 weirdly

0.98

Activations Density 0.312%

punctuation marks and sentence endings

No Comments

No Known Activations