INDEX

Explanations

phrases related to past experiences or states of being

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

ColumnsMode

-0.14

uhan

-0.14

ÑĨÐµ

-0.14

ustria

-0.13

imesteps

-0.13

è¡ĵ

-0.13

Ð°Ð»ÐµÐ¶

-0.13

.lst

-0.13

 Ulus

-0.13

ividual

-0.13

POSITIVE LOGITS

 lately

0.30

 since

0.27

since

0.21

ince

0.20

 recently

0.19

ä»¥æĿ¥

0.18

 Since

0.18

 ÙħÙĨØ°

0.17

Since

0.17

 recent

0.15

Activations Density 0.357%

phrases related to past experiences or states of being

No Comments

No Known Activations