INDEX

Explanations

references to court cases or legal names

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 12-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_12/width_16k/average_l0_82

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.12.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Beſ

-0.86

 Jefus

-0.86

Efq

-0.85

 Theſe

-0.80

Eſ

-0.79

 ſtill

-0.77

Hæ

-0.75

 myſelf

-0.75

AndEndTag

-0.73

umahan

-0.73

POSITIVE LOGITS

fikasi

0.68

asteroide

0.66

wikimedia

0.65

 Normdatei

0.64

defineProperty

0.62

thansa

0.62

TAGS

0.61

von

0.61

TRAS

0.58

 Haas

0.58

Activations Density 0.647%

references to court cases or legal names

No Comments

No Known Activations