INDEX

Explanations

references to time periods and durations, particularly in decades and centuries

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 20-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L20R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.20.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

OrUpdate

-0.18

-0.16

fully

-0.16

-0.15

edom

-0.15

san

-0.15

mund

-0.15

mouth

-0.14

specs

-0.14

POSITIVE LOGITS

-long

0.40

-old

0.28

ago

0.22

-plus

0.21

æľ«

0.21

long

0.20

 boyunca

0.20

icals

0.19

 ÙħÛĮÙĦØ§Ø¯ÛĮ

0.18

-wide

0.18

Activations Density 0.029%

references to time periods and durations, particularly in decades and centuries

No Comments

No Known Activations