INDEX

Explanations

negative phrases or statements

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

å±ĭ

-0.16

isle

-0.16

.MixedReality

-0.15

usterity

-0.15

locate

-0.15

hazi

-0.14

hare

-0.14

itore

-0.14

jee

-0.14

hattan

-0.14

POSITIVE LOGITS

ional

0.30

tingham

0.29

icias

0.26

epad

0.24

ices

0.22

urnal

0.22

 sure

0.21

ori

0.20

 Sure

0.20

arial

0.20

Activations Density 0.055%

negative phrases or statements

No Comments

No Known Activations