INDEX

Explanations

legal arguments related to justifications for classifications and the scrutiny applied to them

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

 houſe

-0.60

 Monfieur

-0.57

 pleaſure

-0.53

 Houſe

-0.52

 deſt

-0.51

 purpoſe

-0.51

AlterField

-0.48

 gabbana

-0.48

leſs

-0.48

 ſtate

-0.47

POSITIVE LOGITS

 justification

0.77

 explanation

0.72

 explicación

0.71

 explanations

0.71

 gius

0.70

 justifying

0.68

 explain

0.66

 Begründung

0.66

 justify

0.66

explique

0.65

Activations Density 0.819%

legal arguments related to justifications for classifications and the scrutiny applied to them

No Comments

No Known Activations