INDEX

Explanations

terms related to legal and professional responsibilities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

pei

-0.18

raquo

-0.16

ADER

-0.15

ÑĮÐ¸

-0.15

KER

-0.15

ingers

-0.14

zdy

-0.14

arker

-0.13

AMED

-0.13

agas

-0.13

POSITIVE LOGITS

entiful

0.18

crud

0.15

McN

0.15

-specific

0.15

oser

0.15

izontal

0.14

cea

0.14

kyt

0.14

 opin

0.14

umi

0.14

Activations Density 0.061%

terms related to legal and professional responsibilities

No Comments

No Known Activations