INDEX

Explanations

references to inmates and prisoners

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

PLICATIONS

-0.63

 Kalk

-0.61

 priesthood

-0.61

positories

-0.61

pur

-0.60

ActionMode

-0.59

ValueStyle

-0.59

 Meksiku

-0.59

}$}

-0.58

тельство

-0.57

POSITIVE LOGITS

 STUDENT

1.05

 student

1.03

 Schüler

0.98

 participant

0.96

 athlete

0.96

 participants

0.96

STUDENT

0.95

 learner

0.95

Athlete

0.94

 Participant

0.94

Activations Density 0.504%

references to inmates and prisoners

No Comments

No Known Activations