INDEX

Explanations

references to codes or identifiers related to technical documentation or programming errors

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

 ProtoMessage

-0.52

AndEndTag

-0.45

setVerticalGroup

-0.40

 bandoulière

-0.40

shid

-0.40

bootstrapcdn

-0.39

Bises

-0.39

alnız

-0.39

 enfans

-0.39

 belang

-0.38

POSITIVE LOGITS

 ModelExpression

0.51

execSQL

0.50

 esternos

0.49

ngdoc

0.42

Carriera

0.40

queryInterface

0.38

setDo

0.38

offset

0.36

fog

0.35

 ©️

0.35

Activations Density 0.288%

references to codes or identifiers related to technical documentation or programming errors

No Comments

No Known Activations