INDEX

Explanations

elements or indicators related to hidden states or components in code

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

+#+#

-0.66

Edy

-0.65

fromnode

-0.61

 Monfieur

-0.61

OnClickListener

-0.61

 Jove

-0.60

FailureListener

-0.60

 carreaux

-0.59

цездатний

-0.59

 devoirs

-0.58

POSITIVE LOGITS

 hidden

1.25

hidden

1.10

 Hidden

1.03

Hidden

1.00

 verborgen

0.86

 concealed

0.72

 скры

0.70

HIDDEN

0.68

actionBar

0.68

GOTREF

0.66

Activations Density 0.021%

elements or indicators related to hidden states or components in code

No Comments

No Known Activations