INDEX

Explanations

references to included and excluded items or costs in a list or description

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

vio

-0.15

 Cole

-0.15

Cel

-0.15

zeich

-0.14

cap

-0.14

 shade

-0.14

post

-0.14

hdc

-0.14

 ATTRIBUTE

-0.14

 shed

-0.14

POSITIVE LOGITS

erland

0.19

orne

0.18

kov

0.17

 selber

0.15

mine

0.15

.sponge

0.15

DMI

0.15

 bidder

0.14

Advertisement

0.14

 mine

0.14

Activations Density 0.102%

references to included and excluded items or costs in a list or description

No Comments

No Known Activations