INDEX

Explanations

phrases related to striving for improvement or breaking boundaries

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

kt

-0.17

xit

-0.16

arser

-0.16

è¡Ĺéģĵ

-0.15

amo

-0.15

/by

-0.14

bsolute

-0.14

kir

-0.14

etimes

-0.14

avig

-0.14

POSITIVE LOGITS

 limits

0.39

 envelope

0.38

 boundaries

0.36

 Limits

0.35

 envelopes

0.33

 buttons

0.31

limits

0.29

 aside

0.28

Limits

0.28

 envelop

0.28

Activations Density 0.037%

phrases related to striving for improvement or breaking boundaries

No Comments

No Known Activations