INDEX

Explanations

phrases indicating changes in levels or amounts

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

Personendaten

-0.62

thalene

-0.61

 muualla

-0.57

WebElementEntity

-0.55

 Sự

-0.54

SpringRunner

-0.54

|}\

-0.51

 Italijani

-0.51

lexikon

-0.51

schild

-0.51

POSITIVE LOGITS

 diventare

0.67

 atteindre

0.67

 deviennent

0.64

值为

0.63

 become

0.63

.*")]

0.62

atteindre

0.60

 appelez

0.59

 достижения

0.59

InputBorder

0.58

Activations Density 0.279%

phrases indicating changes in levels or amounts

No Comments

No Known Activations