INDEX

Explanations

phrases emphasizing the concept of "any"

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

ä¸ĢäºĽ

-0.18

 Various

-0.17

eless

-0.16

 Things

-0.16

rick

-0.16

rej

-0.15

 quelques

-0.15

 some

-0.15

 Some

-0.15

 SOME

-0.15

POSITIVE LOGITS

/all

0.38

THING

0.34

ones

0.34

 sort

0.33

place

0.32

 kind

0.30

kind

0.30

thin

0.29

ht

0.26

ting

0.24

Activations Density 0.057%

phrases emphasizing the concept of "any"

No Comments

No Known Activations