INDEX

Explanations

references to methodologies in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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»Ð¸ÑĨ

-0.17

lets

-0.17

teen

-0.17

.makeText

-0.15

ween

-0.15

let

-0.15

deen

-0.15

lexer

-0.14

á»įng

-0.14

endor

-0.14

POSITIVE LOGITS

ical

0.35

ological

0.35

ologies

0.32

ically

0.31

ologically

0.29

ICAL

0.24

ology

0.24

ologie

0.23

ologic

0.22

icals

0.20

Activations Density 0.044%

references to methodologies in various contexts

No Comments

No Known Activations