INDEX

Explanations

instances of help or assistance-related terms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 3-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_3/width_16k/average_l0_59

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.3.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

VersionUID

-0.88

ſelves

-0.84

 myſelf

-0.80

 Majefty

-0.77

kuuta

-0.71

IBLIO

-0.70

 houſe

-0.69

ſelf

-0.69

 doubtnut

-0.66

 Houſe

-0.65

POSITIVE LOGITS

 help

1.18

 assist

0.92

 Help

0.81

aid

0.78

help

0.74

 autorytatywna

0.73

 support

0.72

 assistance

0.72

Help

0.71

 помочь

0.66

Activations Density 0.071%

instances of help or assistance-related terms

No Comments

No Known Activations