INDEX

Explanations

instances of the word "become" and its variations, indicating transformations or changes in state or status

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

 Often

-0.16

nable

-0.15

ouro

-0.15

thora

-0.14

:nth

-0.14

 often

-0.14

tÃŃ

-0.14

uffy

-0.14

uell

-0.14

ulle

-0.14

POSITIVE LOGITS

 part

0.24

 increasingly

0.23

 known

0.21

 acquainted

0.20

 aware

0.20

 lodged

0.20

 synonymous

0.20

 something

0.20

 extinct

0.19

 involved

0.18

Activations Density 0.083%

instances of the word "become" and its variations, indicating transformations or changes in state or status

No Comments

No Known Activations