INDEX

Explanations

references to specific individuals and their actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

uling

-0.15

outil

-0.14

(contents

-0.14

 dÅĻev

-0.14

abb

-0.13

dust

-0.13

 Ð¼Ð¾Ð±

-0.13

thinkable

-0.13

 Ð·Ð°Ð²Ð¸ÑģÐ¸Ð¼

-0.13

mobile

-0.13

POSITIVE LOGITS

 conced

0.20

oct

0.19

 concede

0.18

oct

0.18

 attrib

0.15

 dedic

0.15

 attributed

0.14

 surrendered

0.14

 Reserve

0.14

vement

0.14

Activations Density 0.020%

references to specific individuals and their actions

No Comments

No Known Activations