INDEX

Explanations

instances of attribution or statements made by individuals

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 2-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L2R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.2.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

ÉĻ

-0.17

 dias

-0.15

Ð¾ÑĪ

-0.14

eren

-0.14

hab

-0.14

ent

-0.14

lands

-0.14

uy

-0.14

 vari

-0.13

POSITIVE LOGITS

ycastle

0.16

ymm

0.15

udic

0.15

ì£¼ìĿĺ

0.15

ffield

0.14

hton

0.14

urma

0.14

elah

0.14

rdf

0.14

-signed

0.14

Activations Density 0.115%

instances of attribution or statements made by individuals

No Known Activations