INDEX

Explanations

variations of the root word "ave" and its related forms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 15-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L15R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.15.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

iff

-0.20

Ã¶m

-0.19

eres

-0.19

wart

-0.18

eron

-0.18

ingu

-0.17

warz

-0.16

alin

-0.16

verbatim

-0.16

uments

-0.15

POSITIVE LOGITS

olution

0.22

illance

0.22

ishly

0.18

alls

0.18

ocation

0.17

ously

0.17

eld

0.17

Ħìŀ¬

0.17

ritable

0.17

antage

0.17

Activations Density 0.649%

variations of the root word "ave" and its related forms

No Comments

No Known Activations