INDEX

Explanations

phrases related to full-length records or albums

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

ÑģÐ¿ÑĸÐ»ÑĮ

-0.15

gio

-0.15

ORIA

-0.15

rips

-0.15

eria

-0.14

lei

-0.14

 noexcept

-0.14

estre

-0.14

esan

-0.14

urer

-0.14

POSITIVE LOGITS

ench

0.17

 concrete

0.15

ook

0.14

zcze

0.14

oltage

0.14

.neo

0.14

iller

0.14

oven

0.14

arming

0.14

_locals

0.14

Activations Density 0.007%

phrases related to full-length records or albums

No Comments

No Known Activations