INDEX

Explanations

phrases indicating health and well-being assessments

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 22-gemmascope-mlp-16k

Configuration

google/gemma-scope-2b-pt-mlp/layer_22/width_16k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.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

\|^{

-0.58

 rör

-0.56

 desar

-0.54

脚注の使い方

-0.52

 Grüsse

-0.52

 generazione

-0.50

 felicità

-0.50

 idade

-0.50

 rayures

-0.50

 ingår

-0.49

POSITIVE LOGITS

FunctionFlags

0.66

 placed

0.61

 Placed

0.59

Portale

0.56

 transfieras

0.56

placed

0.51

 placing

0.50

 placement

0.50

 horseshoe

0.49

 internetowa

0.48

Activations Density 0.269%

phrases indicating health and well-being assessments

No Comments

No Known Activations