INDEX

Explanations

instances of the word "become" and its variations, indicating transformation or change

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 23-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L23R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.23.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

rench

-0.16

nable

-0.16

 toujours

-0.14

uffy

-0.14

icide

-0.14

inz

-0.14

patibility

-0.13

rible

-0.13

utto

-0.13

 sebou

-0.13

POSITIVE LOGITS

 increasingly

0.23

 aware

0.20

 part

0.20

 lodged

0.18

 extinct

0.17

 acquainted

0.16

aware

0.16

ingly

0.16

 apparent

0.15

ate

0.15

Activations Density 0.080%

instances of the word "become" and its variations, indicating transformation or change

No Comments

No Known Activations