INDEX

Explanations

numeric values, including dates and percentages

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 23-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L23R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.23.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

erot

-0.16

ousse

-0.15

zs

-0.15

ensch

-0.15

jin

-0.15

 Ø§ÙĦØ¬ÙĨ

-0.14

irting

-0.14

leur

-0.14

pha

-0.13

 Inspiration

-0.13

POSITIVE LOGITS

iga

0.15

profil

0.15

thren

0.15

odule

0.15

è³

0.14

 Flying

0.14

erras

0.14

ãģĨãģ¡

0.14

izi

0.14

aire

0.14

Activations Density 0.097%

numeric values, including dates and percentages

No Comments

No Known Activations