INDEX

Explanations

punctuation marks and sentence boundaries

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

 Undo

-0.15

ogue

-0.15

Ra

-0.15

dra

-0.14

enz

-0.14

Ra

-0.14

as

-0.14

AIN

-0.14

bill

-0.14

´

-0.14

POSITIVE LOGITS

echa

0.17

ekli

0.16

Ã«l

0.15

elman

0.15

/Dk

0.14

/******/

0.14

bidden

0.14

abled

0.14

 Norris

0.13

ysi

0.13

Activations Density 1.092%

punctuation marks and sentence boundaries

No Comments

No Known Activations