INDEX

Explanations

references to social media and news reporting

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 17-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L17R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.17.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

iferay

-0.20

Ø§ÙĨÙĩ

-0.16

piler

-0.16

ÏģÏħ

-0.15

ingles

-0.15

vsp

-0.15

orge

-0.14

inue

-0.14

awah

-0.14

uggle

-0.14

POSITIVE LOGITS

 Chair

0.15

0.14

ak

0.14

sek

0.14

.pg

0.14

pg

0.14

ÑĢÐ¾Ð¸Ð·

0.14

undi

0.13

 Chairs

0.13

Activations Density 0.064%

references to social media and news reporting

No Comments

No Known Activations