INDEX

Explanations

references to specific dates, events, or notable experiences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

rungsseite

-0.54

 ujednoznacz

-0.46

 Autorisations

-0.45

Cyfeiriadau

-0.44

 autorytatywna

-0.43

 Konami

-0.41

 صوتيه

-0.40

manufact

-0.40

Referanser

-0.40

 Infórmanos

-0.40

POSITIVE LOGITS

 reading

0.63

 read

0.58

 binge

0.54

Reading

0.52

 Reading

0.49

 reads

0.48

 lectura

0.48

 leitura

0.47

reading

0.45

阅读

0.43

Activations Density 0.352%

references to specific dates, events, or notable experiences

No Comments

No Known Activations