INDEX

Explanations

words indicating states of being and existence

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

 moot

-0.15

 Something

-0.14

íķŃ

-0.14

 sight

-0.14

annon

-0.14

inta

-0.14

 anything

-0.13

rench

-0.13

lob

-0.13

itez

-0.13

POSITIVE LOGITS

 happening

0.37

 happen

0.25

 happened

0.24

 happens

0.24

 Happ

0.22

 happ

0.20

åıĳçĶŁ

0.19

 aconte

0.18

wrong

0.17

 wrong

0.17

Activations Density 0.080%

words indicating states of being and existence

No Comments

No Known Activations