INDEX

Explanations

elements in a structured or formatted context, indicating relationships or processes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 35-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_35/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.35.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

 Савезне

-0.54

Monday

-0.51

 estekak

-0.50

 nakalista

-0.50

writeFieldEnd

-0.49

 surla

-0.47

TokenNameLPAREN

-0.46

Referanser

-0.46

 autorytatywna

-0.45

جمعیت

-0.45

POSITIVE LOGITS

⑷

0.68

0.53

 Thursday

0.44

 nargin

0.43

0.41

 konz

0.40

commun

0.39

od

0.39

 Correia

0.39

ⓓ

0.39

Activations Density 0.446%

elements in a structured or formatted context, indicating relationships or processes

No Comments

No Known Activations

elements in a structured or formatted context, indicating relationships or processes

No Comments

No Known Activations