INDEX

Explanations

terms related to consistency and efficiency

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

er

-0.40

hoenix

-0.38

[*]

-0.38

े

-0.35

cmdline

-0.35

ER

-0.34

errcode

-0.34

 Sach

-0.33

Aqu

-0.32

ING

-0.32

POSITIVE LOGITS

 transfieras

0.55

 desmotivaciones

0.55

sizeCache

0.54

󠁿

0.54

 مشين

0.54

tonode

0.54

 pinulongan

0.53

transQ

0.52

 getItemCount

0.51

 فريبيس

0.50

Activations Density 0.328%

terms related to consistency and efficiency

No Comments

No Known Activations