INDEX

Explanations

terms related to criminal activity and justice

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

Crime

-0.16

 Crime

-0.15

 criminals

-0.15

ãģĬãĤĬ

-0.15

agon

-0.15

ãĥªãĥ¼ãĤº

-0.15

 crime

-0.15

crime

-0.15

ettle

-0.14

POSITIVE LOGITS

ity

0.29

istics

0.22

ization

0.21

ized

0.21

mente

0.20

izing

0.20

izes

0.20

 Minds

0.19

isation

0.19

 justice

0.19

Activations Density 0.007%

terms related to criminal activity and justice

No Comments

No Known Activations