INDEX

Explanations

frequent use of the article "the" and variations of "a."

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

an

-0.14

irl

-0.14

sville

-0.14

al

-0.13

ler

-0.13

etc

-0.13

Ø¶Ø§ÙĨ

-0.13

uibModal

-0.13

rey

-0.13

lain

-0.13

POSITIVE LOGITS

ifar

0.15

OKIE

0.14

beck

0.14

aida

0.14

igos

0.13

okane

0.13

gili

0.13

AGEMENT

0.13

 nackte

0.13

cid

0.12

Activations Density 0.029%

frequent use of the article "the" and variations of "a."

No Comments

No Known Activations