INDEX

Explanations

the presence of the abbreviation "Pl" associated with 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 @ 31-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.31.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

yssey

-0.23

at

-0.21

asma

-0.20

colm

-0.19

enty

-0.19

atz

-0.19

etch

-0.18

zeÅĪ

-0.17

ocate

-0.17

ants

-0.17

POSITIVE LOGITS

astics

0.26

ummer

0.23

anners

0.21

atts

0.21

aines

0.20

ural

0.19

unge

0.19

umber

0.19

ots

0.19

atoon

0.19

Activations Density 0.006%

the presence of the abbreviation "Pl" associated with various contexts

No Comments

No Known Activations