INDEX

Explanations

structural elements of code, specifically related to class design and memory management

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

rons

-0.15

rette

-0.15

 drain

-0.14

æŁ±

-0.14

æŀ¶

-0.14

dda

-0.14

omy

-0.14

acula

-0.14

 strain

-0.14

 Sach

-0.14

POSITIVE LOGITS

 virtual

0.22

virtual

0.18

.Virtual

0.15

irtual

0.15

antom

0.15

.virtual

0.15

friend

0.14

explicit

0.14

pga

0.14

 explicit

0.14

Activations Density 0.026%

structural elements of code, specifically related to class design and memory management

No Comments

No Known Activations