INDEX

Explanations

instances of conditional language and phrases indicating possibilities or hypotheses

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 6-gemmascope-mlp-16k

Configuration

google/gemma-scope-2b-pt-mlp/layer_6/width_16k/average_l0_133

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.6.hook_mlp_out

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

 doInBackground

-0.85

Espèce

-0.71

rungsseite

-0.70

 hjemme

-0.69

뀐

-0.66

 AppModule

-0.66

 peuples

-0.64

 Sopho

-0.63

 elä

-0.63

 faptul

-0.62

POSITIVE LOGITS

By

0.50

 Carolina

0.49

thag

0.48

日閲覧

0.47

NotBe

0.47

 เป็น

0.45

 relev

0.45

NESOTA

0.44

ALIAN

0.43

 ivelany

0.43

Activations Density 0.306%

instances of conditional language and phrases indicating possibilities or hypotheses

No Comments

No Known Activations