INDEX

Explanations

numerical values representing years or dates

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 2-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L2R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.2.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

orsk

-0.19

ollen

-0.15

osas

-0.15

antan

-0.15

ojis

-0.14

Sey

-0.14

oken

-0.14

agara

-0.14

erokee

-0.14

asse

-0.14

POSITIVE LOGITS

itele

0.15

borg

0.14

sequent

0.14

ÑĢÐ°Ð²Ð¸

0.14

auer

0.14

çŁ

0.13

 kontakte

0.13

igt

0.13

 iconName

0.13

nÄĽm

0.13

Activations Density 0.151%

numerical values representing years or dates

No Comments

No Known Activations