INDEX

Explanations

programming syntax and logical conditions

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

boolean assignment and comparison

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

Filmografia

-0.53

TryDecodeAsNil

-0.46

 يتيمه

-0.46

Presenta

-0.45

tvguidetime

-0.41

 desmotivaciones

-0.39

 Ubicación

-0.38

acijos

-0.38

 الحره

-0.38

 rejoined

-0.38

POSITIVE LOGITS

 arşivlendi

0.47

хьтан

0.47

 Audiodateien

0.34

TemporalType

0.34

ModelSerializer

0.34

capital

0.33

thargy

0.33

didSet

0.33

 alerg

0.33

fieldId

0.33

Activations Density 0.037%

programming syntax and logical conditions

boolean assignment and comparison

No Comments

No Known Activations