INDEX

Explanations

statements related to business performance and evaluation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

uel

-0.16

enan

-0.15

antan

-0.15

zzo

-0.15

opers

-0.14

elerik

-0.14

 voleb

-0.14

erah

-0.13

issy

-0.13

/antlr

-0.13

POSITIVE LOGITS

æľīäºĽ

0.15

 consegu

0.14

rieved

0.13

adam

0.13

 Ports

0.13

oria

0.13

clr

0.13

 Damen

0.13

 majority

0.13

busy

0.13

Activations Density 0.204%

statements related to business performance and evaluation

No Comments

No Known Activations