INDEX

Explanations

names of researchers and their associated works

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

 kasarigan

-0.82

 للمعارف

-0.82

 estekak

-0.80

 queſta

-0.72

tvguidetime

-0.69

 Geſch

-0.61

 EconPapers

-0.61

 &___

-0.60

 ſind

-0.59

 utafitiHapana

-0.57

POSITIVE LOGITS

0.56

and

0.39

https

0.36

0.34

le

0.33

0.31

INSTANCE

0.30

 estancias

0.30

 Fürst

0.30

Activations Density 0.393%

names of researchers and their associated works

No Comments

No Known Activations