INDEX

Explanations

terms related to measurement or magnitude across 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 @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

zelf

-0.19

ernals

-0.17

iates

-0.16

est

-0.16

sell

-0.16

reads

-0.15

urer

-0.14

nings

-0.14

role

-0.14

urers

-0.14

POSITIVE LOGITS

-up

0.24

-down

0.23

able

0.22

ToFit

0.20

-out

0.20

out

0.20

tron

0.19

way

0.17

ardy

0.17

azy

0.17

Activations Density 0.016%

terms related to measurement or magnitude across various contexts

No Comments

No Known Activations