INDEX

Explanations

dates and specific numerical references within the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 28-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L28R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.28.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

other

-0.17

 NONE

-0.16

OTHER

-0.15

others

-0.15

alach

-0.15

 others

-0.15

 OTHER

-0.15

 ÙĦØ¯ÙĬ

-0.15

 anderen

-0.15

Others

-0.14

POSITIVE LOGITS

 representatives

0.20

we

0.19

 during

0.18

during

0.18

 marked

0.17

Represent

0.17

 marks

0.17

marked

0.17

 During

0.16

 thousands

0.15

Activations Density 0.150%

dates and specific numerical references within the text

No Comments

No Known Activations