INDEX

Explanations

proper nouns and names associated with places or significant entities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 10-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_10/width_16k/average_l0_77

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.10.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

pat

-0.63

et

-0.51

Pat

-0.50

quete

-0.50

-0.49

lecht

-0.49

江湖

-0.48

Require

-0.48

dot

-0.48

POSITIVE LOGITS

 pleaſure

0.91

 Majefty

0.87

 houſe

0.85

 ſever

0.81

Efq

0.81

 myſelf

0.80

 themſelves

0.80

 itſelf

0.79

 leſs

0.78

 ſtate

0.77

Activations Density 0.155%

proper nouns and names associated with places or significant entities

No Comments

No Known Activations