INDEX

Explanations

references to awards and recognition of artistic works

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

usz

-0.16

dek

-0.15

untas

-0.15

 touch

-0.15

onen

-0.14

 Mush

-0.14

orz

-0.14

fish

-0.14

eyJ

-0.14

mani

-0.14

POSITIVE LOGITS

hol

0.18

ÑģÑĮ

0.18

 species

0.16

emade

0.16

ISIBLE

0.15

 Species

0.15

zsche

0.15

ohl

0.15

rac

0.14

 pseud

0.14

Activations Density 0.005%

references to awards and recognition of artistic works

No Comments

No Known Activations