INDEX

Explanations

occurrences of numerical values and references to measurements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

chet

-0.14

ii

-0.14

ott

-0.14

ë¡Ģ

-0.13

uan

-0.13

ories

-0.13

á»ķng

-0.13

oun

-0.13

atel

-0.13

POSITIVE LOGITS

erken

0.15

agrid

0.15

-addons

0.15

å°

0.14

Obr

0.14

<typeof

0.14

 Uncomment

0.14

hci

0.14

Ð»Ð°Ð±

0.13

aspers

0.13

Activations Density 0.178%

occurrences of numerical values and references to measurements

No Comments

No Known Activations