INDEX

Explanations

discussions related to criminal justice and rehabilitation

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

Negative Logits

à¸¸à¹Ī

-0.15

Ã©ric

-0.14

 ØµÙĦÙī

-0.14

utar

-0.14

aar

-0.13

_Device

-0.13

_sampler

-0.13

oftware

-0.13

Ð°ÑĤÐ¸

-0.13

 sublic

-0.13

POSITIVE LOGITS

 prison

0.77

 Prison

0.69

 prisoner

0.67

 prisoners

0.65

 jail

0.64

 inmate

0.63

 prisons

0.61

 inmates

0.59

 Jail

0.55

 incarcerated

0.50

Activations Density 0.377%

discussions related to criminal justice and rehabilitation

No Known Activations