INDEX

Explanations

terms related to the concept of "ultra" or extreme qualities

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

pes

-0.16

pci

-0.15

lessly

-0.15

SYS

-0.15

older

-0.15

ÅŁtir

-0.15

/Foundation

-0.15

icher

-0.14

lessness

-0.14

POSITIVE LOGITS

 Violet

0.27

aviolet

0.26

 violet

0.26

-wide

0.26

-low

0.25

asonic

0.24

fast

0.24

fine

0.24

-high

0.23

iolet

0.23

Activations Density 0.005%

terms related to the concept of "ultra" or extreme qualities

No Comments

No Known Activations