INDEX

Explanations

legal terms and references related to criminal charges and allegations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

jang

-0.15

ulas

-0.15

à¤¿à¤¥

-0.14

thon

-0.14

 destin

-0.14

ython

-0.14

arget

-0.14

alue

-0.14

ãĥ©ãĤ¯

-0.14

 viper

-0.13

POSITIVE LOGITS

 crim

0.18

 connected

0.18

 upgraded

0.18

 crimes

0.17

 Connected

0.17

 aggravated

0.16

 possession

0.16

connected

0.16

 facing

0.16

him

0.16

Activations Density 0.032%

legal terms and references related to criminal charges and allegations

No Comments

No Known Activations