INDEX

Explanations

identifiers and numerical codes related to authors or subjects

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

erti

-0.14

ozem

-0.14

IMAL

-0.14

empo

-0.14

ç«

-0.14

ecome

-0.14

elmet

-0.14

 sighting

-0.13

 jint

-0.13

stairs

-0.13

POSITIVE LOGITS

Orc

0.19

elsif

0.17

oad

0.16

ucken

0.15

 máº¯c

0.15

 disclosing

0.14

.xhtml

0.14

à¥įà¤µà¤°

0.14

icha

0.14

ouver

0.13

Activations Density 0.003%

identifiers and numerical codes related to authors or subjects

No Comments

No Known Activations