INDEX

Explanations

references to legal situations involving capital punishment

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

ampie

-0.15

çĹħ

-0.15

 fined

-0.14

 Veter

-0.14

rh

-0.14

 insult

-0.14

plash

-0.14

 feud

-0.14

 mast

-0.13

sue

-0.13

POSITIVE LOGITS

 innocence

0.35

 Innoc

0.35

DNA

0.32

 convictions

0.29

 wrongful

0.29

 innocent

0.28

DNA

0.28

 exon

0.28

inn

0.27

 wrong

0.27

Activations Density 0.028%

references to legal situations involving capital punishment

No Comments

No Known Activations