INDEX

Explanations

terms related to reflection and contemplation

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

liness

-0.18

ISR

-0.17

emd

-0.16

pel

-0.15

pers

-0.15

ednou

-0.15

ruba

-0.15

aleb

-0.14

Ä°S

-0.14

IER

-0.14

POSITIVE LOGITS

 upon

0.35

ively

0.32

 Upon

0.28

.DeepEqual

0.26

Upon

0.24

ors

0.22

iveness

0.22

/ref

0.21

ance

0.21

ivity

0.19

Activations Density 0.025%

terms related to reflection and contemplation

No Comments

No Known Activations