INDEX

Explanations

phrases indicating legal or judicial decisions

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

madol

-0.84

lblNewLabel

-0.84

findpost

-0.74

ValueStyle

-0.74

bakken

-0.72

 Summers

-0.72

beutel

-0.70

enite

-0.69

veled

-0.69

 Padilla

-0.68

POSITIVE LOGITS

theless

1.31

 anúncio

1.12

OSA

0.92

 Niki

0.91

↵↵↵↵

0.91

strptime

0.89

 itſelf

0.88

GTS

0.88

 Clipper

0.88

 synap

0.88

Activations Density 0.138%

phrases indicating legal or judicial decisions

No Comments

No Known Activations