INDEX

Explanations

phrases that emphasize perspective or opinion

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 16-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L16R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.16.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

andon

-0.16

Specifications

-0.15

ikut

-0.14

.overlay

-0.14

aign

-0.14

 queryInterface

-0.14

vice

-0.14

ught

-0.14

Ø§ÛĮØª

-0.13

ÑĢÐ°Ð»ÑĮ

-0.13

POSITIVE LOGITS

nor

0.36

Nor

0.32

Nor

0.25

nor

0.22

 Nope

0.21

Nah

0.20

 Instead

0.20

éĤ£æł·

0.20

 sondern

0.20

Instead

0.19

Activations Density 0.121%

phrases that emphasize perspective or opinion

No Comments

No Known Activations