INDEX

Explanations

references to crime and investigative scenes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

ender

-0.15

chwitz

-0.15

OST

-0.15

lad

-0.14

veillance

-0.14

.populate

-0.13

çľī

-0.13

ÑıÐ¶

-0.13

bil

-0.13

itness

-0.13

POSITIVE LOGITS

plus

0.17

eny

0.17

enaire

0.15

quina

0.15

andan

0.15

anje

0.14

 twisted

0.14

.cljs

0.14

 vedenÃŃ

0.14

 Subscriber

0.14

Activations Density 0.010%

references to crime and investigative scenes

No Comments

No Known Activations