INDEX

Explanations

references to the concept of "natural" and its various related terms

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

estion

-0.18

idel

-0.17

aign

-0.15

 nhau

-0.14

ossible

-0.14

LabelText

-0.14

Ã«n

-0.14

ÙĨØ§ÙħÙĩ

-0.14

eature

-0.14

asts

-0.14

POSITIVE LOGITS

istic

0.26

istically

0.22

fully

0.20

/native

0.17

ized

0.17

mente

0.17

ERSHEY

0.16

arkan

0.15

ised

0.14

Ã¼stÃ¼

0.14

Activations Density 0.035%

references to the concept of "natural" and its various related terms

No Comments

No Known Activations