INDEX

Explanations

references to specific places, entities, and the context of descriptions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 36-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_36/width_16k/average_l0_120

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.36.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

oyn

-0.30

 meur

-0.27

 mété

-0.26

 films

-0.24

 movies

-0.24

 pubs

-0.24

 Please

-0.24

 occhiali

-0.24

 esteve

-0.23

ningss

-0.23

POSITIVE LOGITS

 možnost

0.93

 implications

0.91

 możliwości

0.88

 faptul

0.88

 possibility

0.87

 advantages

0.87

 situation

0.85

possibility

0.85

situation

0.85

 possibilities

0.85

Activations Density 1.900%

references to specific places, entities, and the context of descriptions

No Comments

No Known Activations