INDEX

Explanations

elements related to classification and roles in a hierarchical context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 16-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L16R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.16.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

izon

-0.16

ayi

-0.15

uste

-0.15

cao

-0.15

chas

-0.15

ulong

-0.14

imei

-0.14

ikat

-0.14

agt

-0.14

à¸ķà¸¥

-0.14

POSITIVE LOGITS

 Tato

0.14

VP

0.14

uzzi

0.14

 stick

0.14

rael

0.13

acula

0.13

pell

0.13

Mam

0.13

 GÃ¼n

0.13

emoc

0.13

Activations Density 0.009%

elements related to classification and roles in a hierarchical context

No Comments

No Known Activations