INDEX

Explanations

sorting and filtering options for displayed information

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 25-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L25R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.25.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

icha

-0.14

illion

-0.14

enza

-0.14

agascar

-0.14

oad

-0.14

REEN

-0.14

INDER

-0.13

ordinates

-0.13

esion

-0.13

inder

-0.13

POSITIVE LOGITS

 descending

0.33

asc

0.33

Asc

0.31

Descending

0.31

 ascending

0.30

 DESC

0.30

ASC

0.29

 descend

0.28

 desc

0.28

descending

0.28

Activations Density 0.040%

sorting and filtering options for displayed information

No Comments

No Known Activations