INDEX

Explanations

terms related to efficiency and productivity improvements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 25-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L25R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.25.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

-thirds

-0.17

-Ð·Ð°

-0.16

Øª

-0.15

FFFFFFFF

-0.15

-fashioned

-0.15

foundland

-0.15

 Repos

-0.15

æĹ¦

-0.15

ause

-0.15

psilon

-0.14

POSITIVE LOGITS

çİĩ

0.27

iciency

0.22

icient

0.20

/product

0.19

/e

0.18

eenth

0.16

iciencies

0.16

iy

0.15

ment

0.15

 gains

0.15

Activations Density 0.023%

terms related to efficiency and productivity improvements

No Comments

No Known Activations