INDEX

Explanations

sections detailing methodology and proof-related terms in academic writing

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

Text sections labeled "Methods" or "Proof"

np_acts-logits-general · gemini-2.0-flash

Methods, Proof, Results sections

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.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

iſchen

-0.79

 ddelwed

-0.79

 Weiſe

-0.77

 ſehen

-0.77

<unused8>

-0.76

 imagui

-0.75

஦

-0.75

<unused28>

-0.75

[@BOS@]

-0.75

<unused14>

-0.75

POSITIVE LOGITS

0.62

Autowired

0.42

：

0.40

UnusedPrivate

0.31

ModelAdmin

0.30

TODO

0.30

0.28

énario

0.28

 TODO

0.28

Activations Density 0.395%

sections detailing methodology and proof-related terms in academic writing

Text sections labeled "Methods" or "Proof"

Methods, Proof, Results sections

No Comments

No Known Activations

sections detailing methodology and proof-related terms in academic writing

Text sections labeled "Methods" or "Proof"

Methods, Proof, Results sections

No Comments

No Known Activations