INDEX

Explanations

elements related to evidence, legal concepts, and police references

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 3-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_3/width_16k/average_l0_59

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.3.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

Efq

-1.03

 initComponents

-1.02

 estekak

-1.00

Obrázky

-1.00

 Jefus

-0.99

UnusedPrivate

-0.96

 NSCoder

-0.94

fjspx

-0.93

 ſeveral

-0.93

principalColumn

-0.93

POSITIVE LOGITS

the

0.63

ing

0.57

 transfer

0.55

 them

0.54

 answer

0.52

ING

0.52

 question

0.51

 figure

0.50

</i>

0.49

us

0.48

Activations Density 1.586%

elements related to evidence, legal concepts, and police references

No Comments

No Known Activations