INDEX

Explanations

the word "to" in various contexts

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

orc

-0.16

anchor

-0.15

 Ø¨Ø§Øª

-0.14

ÑĤÐ°Ð±

-0.14

EEP

-0.14

zb

-0.14

 anchor

-0.14

.fm

-0.14

wers

-0.13

ymax

-0.13

POSITIVE LOGITS

attles

0.15

ãĥ¼ãĥĵ

0.14

okedex

0.14

èĥ

0.14

abcdefgh

0.14

fdc

0.14

ëĬĲ

0.14

estic

0.14

interpreter

0.14

ÐºÐ°Ð¼

0.13

Activations Density 0.003%

the word "to" in various contexts

No Comments

No Known Activations