INDEX

Explanations

references to specific proteins and their attributes in scientific contexts

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

Category prefixes

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-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

RegressionTest

-0.47

appoint

-0.47

 Assignee

-0.47

Rohy

-0.46

ElementException

-0.46

ۜ

-0.46

casus

-0.45

 henvisninger

-0.45

CloseOperation

-0.44

monos

-0.44

POSITIVE LOGITS

 disambiguazione

0.38

WriteBarrier

0.36

ContentAsync

0.36

 nakalista

0.35

 Wish

0.35

RTLI

0.35

 ujednoznacz

0.34

ंदीखरीदारी

0.34

">//

0.32

 masing

0.31

Activations Density 0.142%

references to specific proteins and their attributes in scientific contexts

Category prefixes

No Comments

No Known Activations