INDEX

Explanations

references to sources and citations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

eldon

-0.16

typings

-0.16

inou

-0.15

itas

-0.15

-c

-0.15

rug

-0.14

iliz

-0.14

uggle

-0.14

POSITIVE LOGITS

angu

0.15

AndView

0.14

isser

0.14

ìĤ°

0.14

 ÙħØ§Ùĥ

0.13

adiator

0.13

Éµ

0.13

iddet

0.13

ledge

0.13

oyer

0.13

Activations Density 0.003%

references to sources and citations

No Comments

No Known Activations