INDEX

Explanations

statements about incidents or events involving legal matters and their consequences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

-0.15

 cancers

-0.14

 suicides

-0.14

ê·ľ

-0.13

ubes

-0.13

ajo

-0.13

fak

-0.13

instein

-0.13

 exponent

-0.13

POSITIVE LOGITS

 witnesses

0.27

 police

0.24

 preliminary

0.24

 authorities

0.23

 prelim

0.23

 according

0.22

 Prel

0.21

 investigators

0.20

 officers

0.20

 Witnesses

0.20

Activations Density 0.096%

statements about incidents or events involving legal matters and their consequences

No Comments

No Known Activations