INDEX

Explanations

words and phrases indicating understanding, awareness, and valuing connections

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

 nettoy

-0.60

γωγ

-0.58

abestanden

-0.58

Morfologia

-0.57

 OFDb

-0.56

 precisione

-0.54

Efq

-0.53

complexContent

-0.53

junto

-0.51

 fût

-0.51

POSITIVE LOGITS

 thinks

0.66

 believes

0.65

 wants

0.63

原始内容存档于

0.63

 knows

0.61

 Knows

0.59

 understands

0.58

Hrsg

0.54

 recognizes

0.53

APORE

0.52

Activations Density 0.370%

words and phrases indicating understanding, awareness, and valuing connections

No Comments

No Known Activations