INDEX

Explanations

legal terminology related to liability and court rulings

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 31-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.31.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

çĬ¯ç½ª

-0.15

åĪĳ

-0.15

 crime

-0.14

 Arrest

-0.14

 Investigations

-0.14

criminal

-0.13

Crime

-0.13

åºľ

-0.13

crime

-0.13

 lidi

-0.13

POSITIVE LOGITS

 section

0.23

 subsection

0.21

 Â§

0.20

 reversal

0.20

"[

0.20

 plaintiff

0.20

 paragraph

0.20

 Section

0.20

 absent

0.20

 nothing

0.20

Activations Density 0.254%

legal terminology related to liability and court rulings

No Comments

No Known Activations