INDEX

Explanations

concepts related to arguments and debates

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 15-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L15R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.15.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

otel

-0.18

adaki

-0.18

apiro

-0.16

oph

-0.16

Ø´ØªÙĩ

-0.16

plusplus

-0.15

phinx

-0.15

atten

-0.14

ÐºÐ¾Ð³

-0.14

 ç±³

-0.14

POSITIVE LOGITS

oire

0.21

 Casc

0.15

 pitch

0.15

Drv

0.14

indre

0.14

imuth

0.14

 polarization

0.14

 masc

0.14

å´İ

0.13

 topic

0.13

Activations Density 0.152%

concepts related to arguments and debates

No Comments

No Known Activations