INDEX

Explanations

legal terminology related to guilt and innocence

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 3-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_3/width_16k/average_l0_59

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.3.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

 ANCHE

-0.60

ambién

-0.59

hå

-0.57

こうして

-0.55

Referencie

-0.54

jfree

-0.54

Affirmed

-0.54

 habet

-0.53

 quæ

-0.53

 faktisk

-0.53

POSITIVE LOGITS

 unable

0.79

 inability

0.76

 Unable

0.72

 unsatisfactory

0.71

 inadequate

0.70

 invalid

0.67

 Insufficient

0.67

 initComponents

0.66

 Inadequate

0.65

yntaxException

0.64

Activations Density 0.367%

legal terminology related to guilt and innocence

No Comments

No Known Activations