INDEX

Explanations

references to significant historical events or periods

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

Ãºc

-0.17

alam

-0.16

_MO

-0.14

acente

-0.14

irsch

-0.14

oplast

-0.14

kom

-0.14

ite

-0.13

_DIP

-0.13

Ð©

-0.13

POSITIVE LOGITS

iani

0.19

avel

0.15

inia

0.15

lei

0.15

urses

0.14

 Charter

0.14

sel

0.14

ar

0.14

igner

0.14

Activations Density 0.053%

references to significant historical events or periods

No Comments

No Known Activations