INDEX

Explanations

numerical references and calculations related to legal cases

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-mlp-65k

Configuration

google/gemma-scope-2b-pt-mlp/layer_18/width_65k/average_l0_88

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.18.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

 omge

-0.48

 Armies

-0.44

 Exxon

-0.43

]<<"

-0.43

 Abba

-0.42

gilet

-0.42

forty

-0.41

ERÍA

-0.41

 visibilité

-0.41

 Crater

-0.41

POSITIVE LOGITS

########.

0.80

twimg

0.66

Aholisi

0.66

ymce

0.64

+#+#

0.64

 Chwiliwch

0.62

 дописавши

0.61

poel

0.59

AutoScale

0.58

ագրություններ

0.56

Activations Density 0.122%

numerical references and calculations related to legal cases

No Comments

No Known Activations