INDEX

Explanations

specific terms related to objects and their characteristics or categories

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

ennie

-0.20

erman

-0.15

kovÃ½ch

-0.15

ÐµÐ»ÐµÐ½

-0.15

Ð»Ð¸Ð¹

-0.14

pressive

-0.14

Ú¯Ø§Ùĩ

-0.14

ture

-0.14

REAK

-0.14

Ã¡lido

-0.14

POSITIVE LOGITS

Å¯

0.34

Ã³w

0.31

åĢĳ

0.27

ä»¬

0.26

Ð°Ñħ

0.26

Ð°Ð¼Ð¸

0.25

ÑĸÐ²

0.24

Ñĭ

0.23

Å¯m

0.22

Ð°Ð¼

0.22

Activations Density 0.050%

specific terms related to objects and their characteristics or categories

No Comments

No Known Activations