INDEX

Explanations

occurrences of the term "La" as a significant reference

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

 étoient

-1.13

 avoient

-1.09

parsedMessage

-1.09

 romantique

-1.08

 feroit

-1.07

 ainfi

-1.04

 provisoire

-1.03

Datuak

-1.01

 Monfieur

-1.01

/*---

-1.00

POSITIVE LOGITS

La

2.04

La

1.92

la

1.88

LA

1.43

la

1.39

LA

1.34

 Laur

1.18

 ла

1.17

 LaSalle

1.11

Ла

1.08

Activations Density 0.049%

occurrences of the term "La" as a significant reference

No Comments

No Known Activations