INDEX

Explanations

text related to legal terminology and notable cases

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 41-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_41/width_16k/average_l0_113

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.41.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ins

-0.35

 potentially

-0.33

 quite

-0.29

 relative

-0.29

 ever

-0.28

 consequently

-0.27

 surprisingly

-0.26

 tầm

-0.25

 obviously

-0.25

 explicitly

-0.25

POSITIVE LOGITS

 незавершена

0.73

AndEndTag

0.69



0.66

 indígen

0.65

JAKARTA

0.63

TypedDataSet

0.62

 increí

0.62

JsonPropertyName

0.61

üyada

0.60

enterOuterAlt

0.60

Activations Density 0.489%

text related to legal terminology and notable cases

No Comments

No Known Activations