INDEX

Explanations

occurrences of the word "push" and its variations

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Ð¸ÑĩÐ½Ð¾

-0.17

adle

-0.16

utsch

-0.16

eker

-0.15

.Xaml

-0.15

RESS

-0.15

ãģıãĤī

-0.15

 Ð¸ÑģÐ¿Ð¾Ð»Ð½

-0.14

jac

-0.14

stown

-0.14

POSITIVE LOGITS

(push

0.28

.Push

0.26

 push

0.25

-push

0.24

 Push

0.22

push

0.21

pull

0.21

Push

0.21

 pushing

0.21

.push

0.21

Activations Density 0.027%

occurrences of the word "push" and its variations

No Comments

No Known Activations