INDEX

Explanations

phrases that indicate a lack of novelty or recycling of ideas

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

onis

-0.19

ogie

-0.19

ONO

-0.16

lÃŃ

-0.16

inke

-0.16

TEMPL

-0.15

Gow

-0.14

ÐµÑİ

-0.14

vro

-0.14

'gc

-0.14

POSITIVE LOGITS

 nothing

0.36

 surprise

0.31

nothing

0.29

 Surprise

0.29

 Nothing

0.28

new

0.26

 NOTHING

0.26

 novel

0.26

Nothing

0.24

 surprises

0.24

Activations Density 0.102%

phrases that indicate a lack of novelty or recycling of ideas

No Comments

No Known Activations