INDEX

Explanations

phrases indicating uncertainty or conditionals

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

almost

-0.75

 almost

-0.73

 autorytatywna

-0.73

 virtually

-0.73

 Virtually

-0.58

 presque

-0.58

cticut

-0.56

Almost

-0.56

GIVEREF

-0.55

 onFailure

-0.54

POSITIVE LOGITS

 necessariamente

0.84

 necessarily

0.83

 necesariamente

0.81

 nécessairement

0.74

 Tampoco

0.73

 forcément

0.73

nor

0.69

 anymore

0.69

necessarily

0.69

 Anſ

0.68

Activations Density 0.230%

phrases indicating uncertainty or conditionals

No Comments

No Known Activations