INDEX

Explanations

concepts related to memory and consciousness

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

éĩĩçĶ¨

-0.15

mart

-0.14

endas

-0.14

metis

-0.14

ongyang

-0.14

sville

-0.13

DAT

-0.13

ugu

-0.13

DI

-0.13

pecially

-0.13

POSITIVE LOGITS

 proper

0.21

 experienced

0.18

 Proper

0.18

 itself

0.18

 impressed

0.18

 Substance

0.18

 substance

0.17

 tout

0.16

 Reality

0.16

 according

0.16

Activations Density 0.407%

concepts related to memory and consciousness

No Comments

No Known Activations