INDEX

Explanations

terms related to legal processes and regulatory decisions

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

hops

-0.56

phens

-0.50

hop

-0.47

 nhiêu

-0.46

 volon

-0.45

ünstig

-0.45

 hoje

-0.44

how

-0.41

НИК

-0.41

denly

-0.40

POSITIVE LOGITS

drawal

0.97

Efq

0.92

 itſelf

0.91

SOUNDBITE

0.89

 Theſe

0.86

 ―――――

0.86

Eſ

0.85

 متعلقه

0.84

titure

0.82

 Monfieur

0.82

Activations Density 1.353%

terms related to legal processes and regulatory decisions

No Comments

No Known Activations