INDEX

Explanations

programming terminology and syntax

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 13-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L13R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.13.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

Î²Î¬

-0.17

utos

-0.15

wyn

-0.15

 Morgan

-0.15

fos

-0.14

 sound

-0.14

İ

-0.13

uto

-0.13

ARGET

-0.13

cci

-0.13

POSITIVE LOGITS

chop

0.16

ubber

0.15

 Pett

0.15

brick

0.14

imus

0.14

readystatechange

0.14

SSF

0.14

alsa

0.14

ovolta

0.13

Equip

0.13

Activations Density 0.034%

programming terminology and syntax

No Comments

No Known Activations