INDEX

Explanations

terminology related to winning, especially in competitive contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 13-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L13R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.13.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

enties

-0.20

cki

-0.17

itize

-0.16

oppins

-0.16

-0.15

edriver

-0.15

vore

-0.15

omatic

-0.15

serv

-0.14

osate

-0.14

POSITIVE LOGITS

elm

0.19

nik

0.19

ively

0.18

een

0.18

inel

0.17

iful

0.17

drink

0.16

minster

0.15

stroke

0.15

rice

0.14

Activations Density 0.064%

terminology related to winning, especially in competitive contexts

No Comments

No Known Activations