INDEX

Explanations

location and event names, particularly related to India

oai_token-act-pair · gemini-2.0-flash

names and symbols related to Ayodhya

np_token-act-pair-logits · gpt-4o-mini

India related context

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_21/width_16k/average_l0_13

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.ln2.hook_normalized

Architecture

jumprelu_transcoder

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

 تانيه

-0.85

 Theſe

-0.78

Efq

-0.77

 myſelf

-0.76

 raiſ

-0.73

 Houſe

-0.71

 itſelf

-0.71

 fevere

-0.71

 stiefel

-0.69

 faſt

-0.69

POSITIVE LOGITS

Sw

0.73

sw

0.64

At

0.60

Sw

0.59

 skill

0.59

 Skill

0.57

Ay

0.56

Am

0.53

0.52

Sag

0.51

Activations Density 0.685%

location and event names, particularly related to India

names and symbols related to Ayodhya

India related context

No Known Activations