INDEX

Explanations

references to investigative processes and personnel related to legal matters

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

illin

-0.16

æ¾¤

-0.15

.ng

-0.14

Care

-0.14

enti

-0.14

DAQ

-0.14

 pstmt

-0.14

housing

-0.13

izer

-0.13

ipple

-0.13

POSITIVE LOGITS

 private

0.32

 Private

0.30

 investigator

0.28

 PRIVATE

0.28

 investigators

0.28

Private

0.27

private

0.26

PRIVATE

0.25

PI

0.25

 Investigators

0.25

Activations Density 0.008%

references to investigative processes and personnel related to legal matters

No Comments

No Known Activations