INDEX

Explanations

references to titles, names, and designations typically found in academic or formal contexts

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

Two-letter prefixes

np_acts-logits-general · gemini-2.0-flash

initial letter combinations followed by words

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.18.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

ed

-0.70

اث

-0.62

ename

-0.60

es

-0.59

lime

-0.58

ease

-0.57

ila

-0.56

ex

-0.56

aste

-0.56

ew

-0.54

POSITIVE LOGITS

 חיצוניים

0.77

RRRR

0.71

rrggbb

0.68

 kasarigan

0.66

IsInitialized

0.66

IsMutable

0.65

LEncoder

0.65

Xr

0.64

TIR

0.63

Activations Density 0.456%

references to titles, names, and designations typically found in academic or formal contexts

Two-letter prefixes

initial letter combinations followed by words

No Comments

No Known Activations