INDEX

Explanations

references to holders or holding mechanisms in various contexts

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

"holder" (or variations) after certain words

np_acts-logits-general · gemini-2.0-flash

viewholder or holder

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

-0.44

تقاوى

-0.40

iostream

-0.40

 Audiodateien

-0.40

lodash

-0.38

stdc

-0.38

 Taft

-0.38

ver

-0.37

hetamine

-0.37

￡

-0.37

POSITIVE LOGITS

ViewHolder

3.42

 ViewHolder

2.22

holder

1.48

 viewHolder

1.45

Holder

1.45

 Holder

1.30

 holder

1.28

holders

1.27

viewHolder

1.24

HOLDER

1.23

Activations Density 0.056%

references to holders or holding mechanisms in various contexts

"holder" (or variations) after certain words

viewholder or holder

No Comments

No Known Activations