INDEX

Explanations

instances of dialogue or quotes from users in a discussion

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

 infl

-0.15

otta

-0.15

aman

-0.14

iva

-0.14

uil

-0.14

 Sharp

-0.14

á»ĳt

-0.14

emer

-0.14

 Univ

-0.14

ë¹

-0.13

POSITIVE LOGITS

abler

0.20

undles

0.16

groupid

0.15

tures

0.15

\OptionsResolver

0.15

ettings

0.14

ë²Ī

0.14

ILER

0.14

istar

0.14

á»ĳi

0.14

Activations Density 0.008%

instances of dialogue or quotes from users in a discussion

No Comments

No Known Activations