INDEX

Explanations

references to copyright and legal permissions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 17-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L17R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.17.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

uche

-0.16

 Tham

-0.16

rim

-0.15

itional

-0.14

issan

-0.14

esen

-0.14

speaker

-0.14

kf

-0.13

ujet

-0.13

POSITIVE LOGITS

verbatim

0.17

AW

0.16

Copying

0.15

ãĥ³ãĥĸ

0.15

allas

0.14

 cred

0.14

/Foundation

0.14

/tab

0.14

tero

0.14

reme

0.14

Activations Density 0.034%

references to copyright and legal permissions

No Comments

No Known Activations