INDEX

Explanations

occurrences of opening or closing tags in various kinds of documents

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

 blanches

-0.78

 atque

-0.70

 obicei

-0.69

 nimic

-0.67

 dítě

-0.67

 nemlig

-0.64

 negru

-0.64

 stycz

-0.63

 sanitaires

-0.63

 culoare

-0.62

POSITIVE LOGITS

 Dari

0.96

UrlResolution

0.90

zaki

0.84

Dari

0.83

ropriate

0.82

là

0.82

={`/

0.79

 незавершена

0.77

ilà

0.77

 của

0.77

Activations Density 0.013%

occurrences of opening or closing tags in various kinds of documents

No Comments

No Known Activations