INDEX

Explanations

terms related to evidence and legal context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 4-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_4/width_16k/average_l0_124

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.4.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

principalColumn

-0.79

WebVitals

-0.79

ConstraintMaker

-0.78

ArrowToggle

-0.76

Autoritní

-0.74

 MainAxisSize

-0.74

URLException

-0.74

uteurs

-0.73

parsedMessage

-0.72

expandindo

-0.70

POSITIVE LOGITS

 evidence

0.86

 footage

0.80

 stuff

0.79

stuff

0.74

 equipment

0.74

evidence

0.73

 legislation

0.73

 advice

0.73

 information

0.72

 gear

0.71

Activations Density 3.493%

terms related to evidence and legal context

No Comments

No Known Activations