INDEX

Explanations

terms related to layout and positioning in programming or design contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 20-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L20R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

gent

-0.16

ë¨¹

-0.16

pler

-0.16

inu

-0.15

owing

-0.15

ansson

-0.15

opers

-0.15

night

-0.15

Hao

-0.15

ACE

-0.15

POSITIVE LOGITS

.createSequentialGroup

0.23

.createParallelGroup

0.22

(Layout

0.20

kowski

0.18

Subviews

0.18

weise

0.17

/Layout

0.17

hift

0.17

.setVerticalGroup

0.17

istik

0.16

Activations Density 0.010%

terms related to layout and positioning in programming or design contexts

No Comments

No Known Activations