INDEX

Explanations

specific brand names and high-value items

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L1R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.1.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

olet

-0.16

ione

-0.15

anch

-0.15

Ð¾Ð±Ð¸

-0.15

ìĦ±

-0.14

ÏģÎ¿Î½

-0.14

rub

-0.14

ÏģÏī

-0.14

/*@

-0.14

ief

-0.14

POSITIVE LOGITS

iros

0.15

 Hava

0.15

irm

0.15

 Mend

0.15

éĸĢ

0.15

ãĥ¬ãĥ³

0.14

vertime

0.14

 ponto

0.14

Activations Density 0.272%

specific brand names and high-value items

No Comments

No Known Activations