INDEX

Explanations

references to increasing online engagement or traffic

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

 futures

-0.15

oyal

-0.15

acha

-0.15

 Ð½Ð°Ñģ

-0.14

utures

-0.14

RAIN

-0.14

obel

-0.14

TU

-0.14

engkap

-0.14

allas

-0.13

POSITIVE LOGITS

 lately

0.26

 newfound

0.20

 increased

0.20

 recently

0.19

æľĢè¿ĳ

0.18

 increasingly

0.18

 Increased

0.17

 recent

0.17

Increased

0.16

.mo

0.16

Activations Density 0.030%

references to increasing online engagement or traffic

No Comments

No Known Activations