INDEX

Explanations

variations of the word "have."

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

"have to"

np_acts-logits-general · gemini-2.0-flash

have to

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-9b-it-res/layer_9/width_16k/average_l0_88

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.9.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

 équilibr

-0.45

 jamás

-0.38

 salga

-0.32

 pluie

-0.31

Protocols

-0.30

 Grünen

-0.30

 azért

-0.30

 sonrisa

-0.30

 paja

-0.30

 alivio

-0.29

POSITIVE LOGITS

 zwiſchen

0.69

 Geiſt

0.66

 imagui

0.65

RenderAtEndOf

0.63

<unused47>

0.63

<unused43>

0.63

 ſeine

0.63

 ſeyn

0.63

<unused3>

0.63

<unused8>

0.63

Activations Density 0.238%

variations of the word "have."

"have to"

have to

No Comments

No Known Activations