INDEX

Explanations

numerical data and statistics related to assessments or applications

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 13-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L13R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.13.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

icho

-0.18

orado

-0.16

'gc

-0.15

 HEAP

-0.15

ascus

-0.15

ustos

-0.15

ypo

-0.15

arrant

-0.15

chy

-0.14

lyph

-0.14

POSITIVE LOGITS

 already

0.20

å·²

0.17

 Already

0.16

core

0.16

lah

0.16

Already

0.16

å·²ç»ı

0.16

å°½

0.15

already

0.15

 Partial

0.15

Activations Density 0.100%

numerical data and statistics related to assessments or applications

No Comments

No Known Activations