INDEX

Explanations

references to invitations or calls to action

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

azzo

-0.16

eness

-0.16

pio

-0.16

eron

-0.15

peration

-0.15

Å¾ÃŃ

-0.15

.ep

-0.15

ugal

-0.14

epam

-0.14

POSITIVE LOGITS

inv

0.30

Inv

0.26

olved

0.26

incible

0.25

inv

0.25

TargetException

0.25

(inv

0.24

.inv

0.23

aders

0.22

-inv

0.22

Activations Density 0.013%

references to invitations or calls to action

No Comments

No Known Activations