INDEX

Explanations

references to structured data or classifications involving "Mat" and "P."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

ÐµÑĢÐ±

-0.16

thin

-0.15

 thin

-0.15

 Thin

-0.15

agra

-0.15

ulary

-0.15

hurst

-0.15

Ð¾Ð¿ÑĢÐ¸

-0.14

ìĪĺê°Ģ

-0.14

Lem

-0.13

POSITIVE LOGITS

llib

0.18

0.15

Ð¾Ð·Ð¸

0.15

Ä±ÅŁÄ±k

0.14

 voksne

0.14

Î³Îµ

0.14

cko

0.14

lli

0.14

Activations Density 0.020%

references to structured data or classifications involving "Mat" and "P."

No Comments

No Known Activations