INDEX

Explanations

references to crime and legal proceedings

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 28-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L28R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.28.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

ijn

-0.16

.gov

-0.16

ê¸Ī

-0.15

ocre

-0.15

aign

-0.15

è®¿

-0.14

ÄĽst

-0.14

ä»ĭ

-0.14

 damer

-0.14

 sideline

-0.14

POSITIVE LOGITS

 faces

0.38

ple

0.37

ple

0.37

Ple

0.33

 plead

0.31

 pleaded

0.30

 pled

0.30

 pleading

0.30

 facing

0.29

 face

0.28

Activations Density 0.098%

references to crime and legal proceedings

No Comments

No Known Activations