INDEX

Explanations

acronyms or initialisms related to organizations, scientific studies, or terminology

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

EMENT

-0.18

ipment

-0.17

ECH

-0.16

akis

-0.16

UREMENT

-0.15

immel

-0.15

anan

-0.15

uli

-0.15

STALL

-0.15

ment

-0.14

POSITIVE LOGITS

dzi

0.17

rze

0.17

AM

0.16

ewis

0.16

erken

0.16

entifier

0.15

ynchronize

0.15

yth

0.15

eri

0.15

iki

0.15

Activations Density 0.244%

acronyms or initialisms related to organizations, scientific studies, or terminology

No Comments

No Known Activations