INDEX

Explanations

technical terms and code structure related to programming

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 25-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L25R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.25.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

gzip

-0.16

ollo

-0.16

 Pere

-0.15

astle

-0.14

Ð°ÑĢÑı

-0.14

anes

-0.14

Knight

-0.13

 Knight

-0.13

 thÃłnh

-0.13

POSITIVE LOGITS

lav

0.20

lav

0.19

PTS

0.19

nal

0.18

av

0.18

InParameter

0.17

nero

0.17

bic

0.17

AV

0.16

avn

0.16

Activations Density 0.022%

technical terms and code structure related to programming

No Comments

No Known Activations