INDEX

Explanations

numerical values related to time 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 @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

erre

-0.17

Ð¾Ð¶

-0.16

seau

-0.15

ilece

-0.14

otre

-0.14

velt

-0.14

 çĨ

-0.14

illow

-0.14

ãģŀ

-0.14

@student

-0.13

POSITIVE LOGITS

å¹´

0.31

ëħĦ

0.28

 å¹´

0.22

0.21

-MM

0.19

-present

0.17

âĢĳ

0.17

Activations Density 0.044%

numerical values related to time or dates

No Comments

No Known Activations