INDEX

Explanations

words associated with academic or formal writing

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L8R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.8.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

aira

-0.18

-0.15

ulen

-0.15

 Isle

-0.15

 behalf

-0.14

etch

-0.14

ialized

-0.14

apÄ±

-0.14

rnek

-0.14

ught

-0.14

POSITIVE LOGITS

anky

0.15

assa

0.15

asser

0.15

 Grat

0.15

avia

0.15

Coc

0.14

ÏīÏĤ

0.14

çĭ¬

0.13

iele

0.13

egg

0.13

Activations Density 0.003%

words associated with academic or formal writing

No Comments

No Known Activations