INDEX

Explanations

references to academic publications and identification of authors

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

Dud

-0.15

 Primer

-0.14

uy

-0.14

spb

-0.14

exact

-0.14

rys

-0.14

ÑĢÐ¾ÑģÑĤÐ¾

-0.14

icy

-0.14

linger

-0.14

ughter

-0.13

POSITIVE LOGITS

ÙĬÙĨØ©

0.15

cko

0.14

 promin

0.13

APT

0.13

ngo

0.13

 Cros

0.13

Ļ

0.13

lilik

0.13

.webkit

0.13

abis

0.13

Activations Density 0.008%

references to academic publications and identification of authors

No Comments

No Known Activations