INDEX

Explanations

phrases related to restoration, relocation, or revisiting concepts and actions

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

degradable

-0.58

modb

-0.56

choly

-0.54

anskje

-0.52

 kasarigan

-0.52

beeren

-0.52

quently

-0.52

altri

-0.52

ليج

-0.51

parency

-0.50

POSITIVE LOGITS

 returning

0.91

 rejoined

0.90

 regain

0.88

 regaining

0.88

 returned

0.88

 Returning

0.88

に戻

0.87

wieder

0.85

 wieder

0.85

戻

0.84

Activations Density 0.495%

phrases related to restoration, relocation, or revisiting concepts and actions

No Comments

No Known Activations