INDEX

Explanations

words and phrases related to legal terms and processes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.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

PerformLayout

-0.74

 impactful

-0.72

sizeCache

-0.64

LEVANT

-0.59

 playbook

-0.58

webElementXpaths

-0.58

 kasarigan

-0.57

 incentiv

-0.57

 transitioning

-0.56

 clunky

-0.56

POSITIVE LOGITS

 eenig

0.76

 Надо

0.74

 daß

0.74

faßt

0.73

 thence

0.71

 gewiß

0.69

 müßte

0.68

 quelquefois

0.68

 muß

0.68

 mußten

0.68

Activations Density 0.633%

words and phrases related to legal terms and processes

No Comments

No Known Activations