INDEX

Explanations

references to geopolitical conflicts, particularly involving Russia, North Korea, and related political actors

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 10-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_10/width_16k/average_l0_77

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.10.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

TestTools

-0.61

TagMode

-0.53

:✨

-0.52

 esboço

-0.50

Autoritní

-0.50

 terecht

-0.50

 jambes

-0.50

 superposition

-0.49

 crutches

-0.48

 esempi

-0.48

POSITIVE LOGITS

 Jinping

0.60

 Familienname

0.60

 flag

0.54

 île

0.53

tain

0.51

nto

0.51

medriver

0.50

 success

0.50

ex

0.50

aguay

0.50

Activations Density 0.182%

references to geopolitical conflicts, particularly involving Russia, North Korea, and related political actors

No Comments

No Known Activations