INDEX

Explanations

references to multimedia content or promotions

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

 sÃ¡t

-0.19

Invocation

-0.15

apos

-0.15

ulton

-0.14

udeau

-0.14

rrha

-0.14

 Fres

-0.14

mos

-0.13

 Sikh

-0.13

ibus

-0.13

POSITIVE LOGITS

elper

0.16

 Gesture

0.14

 Shan

0.14

oid

0.14

otime

0.14

eting

0.14

iali

0.14

.Helper

0.14

 Combination

0.13

áº§m

0.13

Activations Density 0.004%

references to multimedia content or promotions

No Comments

No Known Activations