INDEX

Explanations

references to specific character names or abilities within a fictional context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 0-gemmascope-mlp-65k

Configuration

google/gemma-scope-2b-pt-mlp/layer_0/width_65k/average_l0_72

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.0.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

-0.66

-0.60

ा

-0.60

Jew

-0.57

-0.55

IPC

-0.52

षण

-0.52

LayoutConstraint

-0.50

enumii

-0.50

IBA

-0.50

POSITIVE LOGITS

 Meksiku

0.79

'],

0.74

Espèce

0.73

Rüyada

0.73

 boî

0.73

'])

0.72

 touristes

0.72

 épaules

0.70

]));

0.69

%%

0.69

Activations Density 0.023%

references to specific character names or abilities within a fictional context

No Comments

No Known Activations