INDEX

Explanations

common words like articles ("a", "the"), prepositions ("of"), and words indicating comparison ("like", "more").

oai_token-act-pair · gemini-2.0-flash

a

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_22/width_16k/average_l0_15

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.ln2.hook_normalized

Architecture

jumprelu_transcoder

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 difficulté

-0.58

 détermination

-0.53

 sonno

-0.50

 quantités

-0.48

 délais

-0.48

Vea

-0.47

 déterminé

-0.47

 spesa

-0.46

seamnă

-0.45

 ejemplar

-0.45

POSITIVE LOGITS

twimg

0.82

Scénario

0.70

Espèce

0.69

béco

0.66

 AssemblyTitle

0.66

secta

0.61

fromnode

0.60

Solución

0.60

 سكانية

0.60

Хьажоргаш

0.59

Activations Density 3.655%

common words like articles ("a", "the"), prepositions ("of"), and words indicating comparison ("like", "more").

a

No Comments

No Known Activations