INDEX

Explanations

numerical data and statistics

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

nel

-0.17

-ÑĤÐ¾

-0.17

EDI

-0.15

ffi

-0.15

ÙĬØ§ÙĨ

-0.15

ÐµÐ½Ð½Ñĸ

-0.15

han

-0.14

kel

-0.14

rub

-0.14

 Ð²Ð¾Ð²

-0.14

POSITIVE LOGITS

ilers

0.19

ceeded

0.17

Ã¤s

0.15

unce

0.14

sah

0.14

ers

0.14

Cue

0.14

agu

0.14

jabi

0.14

 nonatomic

0.14

Activations Density 0.067%

numerical data and statistics

No Comments

No Known Activations