INDEX

Explanations

various comments or annotations within the code

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

gs

-0.17

anni

-0.17

icz

-0.15

rase

-0.15

enders

-0.14

rak

-0.14

ut

-0.14

 Jamal

-0.13

GS

-0.13

lo

-0.13

POSITIVE LOGITS

OLOR

0.17

oret

0.16

agnar

0.15

ERRU

0.15

.synthetic

0.14

aghetti

0.14

-pocket

0.14

ë¡±

0.14

.scalablytyped

0.14

lander

0.14

Activations Density 0.049%

various comments or annotations within the code

No Comments

No Known Activations