INDEX

Explanations

programming-related concepts, specifically focusing on class and struct definitions in 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

sko

-0.17

arkan

-0.17

 Equals

-0.16

ë²

-0.15

INCLUDED

-0.15

 Equal

-0.15

idebar

-0.15

.EventQueue

-0.14

lbrace

-0.14

oust

-0.14

POSITIVE LOGITS

busy

0.15

adden

0.15

 busy

0.15

 Progress

0.14

æĤŁ

0.14

 abstract

0.14

arith

0.14

andle

0.13

 furry

0.13

 busiest

0.13

Activations Density 0.002%

programming-related concepts, specifically focusing on class and struct definitions in code

No Comments

No Known Activations