INDEX

Explanations

symbolic indicators and delimiters used in code and textual data

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

apel

-0.15

 conf

-0.15

opt

-0.15

strup

-0.15

Ð¿Ð¸

-0.14

olini

-0.14

inx

-0.14

arrants

-0.14

POSITIVE LOGITS

.scalablytyped

0.16

 jadx

0.15

/navbar

0.15

/socket

0.15

WithType

0.15

 Marketable

0.14

 GroupLayout

0.14

errupted

0.14

.IContainer

0.14

ali

0.14

Activations Density 0.309%

symbolic indicators and delimiters used in code and textual data

No Comments

No Known Activations