INDEX

Explanations

references to software bugs and issues

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 5-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L5R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.5.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

 Held

-0.14

æı

-0.14

rita

-0.14

igm

-0.14

law

-0.14

quires

-0.14

ourse

-0.14

idad

-0.13

ubar

-0.13

OLT

-0.13

POSITIVE LOGITS

lion

0.18

geois

0.16

amet

0.15

Â¤Â¤

0.14

lisi

0.14

killer

0.14

eldorf

0.13

elsif

0.13

ellation

0.13

Fav

0.13

Activations Density 0.015%

references to software bugs and issues

No Comments

No Known Activations