INDEX

Explanations

names and descriptors associated with prominence or significance

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

ness

-0.20

-0.15

é£İ

-0.15

ose

-0.15

kir

-0.14

the

-0.14

/address

-0.14

//

-0.14

POSITIVE LOGITS

 halinde

0.18

Sharper

0.17

edly

0.16

å¼ı

0.16

/template

0.16

/example

0.16

edii

0.15

ãĥ³ãĥĨãĤ£

0.15

çº§

0.15

ishly

0.15

Activations Density 0.196%

names and descriptors associated with prominence or significance

No Comments

No Known Activations