INDEX

Explanations

phrases that indicate agreement and consistency with legal standards or principles

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 7-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_7/width_16k/average_l0_69

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.7.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

 InputDecoration

-0.57

addContainerGap

-0.54

 UserController

-0.52

 gebo

-0.52

expandindo

-0.49

перь

-0.47

Carriera

-0.46

 numberOfRows

-0.45

IsNotNull

-0.45

Scotty

-0.44

POSITIVE LOGITS

Consistent

0.93

 accords

0.89

 contradict

0.88

 contradicts

0.88

 Consistent

0.86

 conforms

0.85

 consistent

0.81

 conform

0.77

 accord

0.77

consistent

0.76

Activations Density 0.418%

phrases that indicate agreement and consistency with legal standards or principles

No Comments

No Known Activations