INDEX

Explanations

terms related to nobility and titles

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Meksiku

-0.56

kere

-0.53

ⓘ

-0.48

realtime

-0.48

তথ্যসূত্র

-0.47

gawai

-0.47

tifact

-0.47

 therefor

-0.47

 труда

-0.45

SPATH

-0.45

POSITIVE LOGITS

 Countess

1.05

 Viscount

1.02

 Baroness

0.96

 Duke

0.94

 marquis

0.93

 Marquis

0.92

 DUKE

0.89

 Duchess

0.86

Efq

0.84

 lords

0.84

Activations Density 0.159%

terms related to nobility and titles

No Comments

No Known Activations