INDEX

Explanations

terms related to software and web technologies

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

³

-0.17

ekil

-0.15

ivent

-0.15

emens

-0.14

Navig

-0.14

luck

-0.14

Fal

-0.14

-www

-0.14

liga

-0.14

.SIG

-0.13

POSITIVE LOGITS

eler

0.15

 revenue

0.14

vetica

0.14

ãĥ¼ãĥª

0.14

Kelly

0.13

tap

0.13

 Kelly

0.13

-chan

0.13

 Early

0.13

adge

0.13

Activations Density 0.026%

terms related to software and web technologies

No Comments

No Known Activations