INDEX

Explanations

terms related to mathematical or computational concepts and operations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

atal

-0.16

Kid

-0.16

Pa

-0.15

æļ

-0.15

Haw

-0.15

ESP

-0.15

 Universal

-0.15

-0.14

Kid

-0.14

POSITIVE LOGITS

CKER

0.15

orque

0.15

áº¯c

0.15

 Ø±Ø³ÙĪÙĦ

0.15

PEND

0.15

inston

0.14

antanamo

0.14

[Math

0.14

.setPrototypeOf

0.14

ziehung

0.14

Activations Density 0.026%

terms related to mathematical or computational concepts and operations

No Comments

No Known Activations