INDEX

Explanations

mathematical terminology and notation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

áº¡o

-0.15

anga

-0.15

ayo

-0.14

æĤł

-0.14

eton

-0.14

_invoke

-0.13

 Brady

-0.13

avin

-0.13

oser

-0.13

istes

-0.13

POSITIVE LOGITS

normal

0.27

 normal

0.25

bf

0.20

up

0.19

Normal

0.19

it

0.18

 Normal

0.18

ormal

0.17

md

0.17

Activations Density 0.022%

mathematical terminology and notation

No Comments

No Known Activations