INDEX

Explanations

references to news sources and article citations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 23-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L23R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.23.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

_ATTRIB

-0.15

íĿ¬

-0.15

isci

-0.14

 Î±ÏĨ

-0.14

Î¼ÏĨ

-0.14

 Ð²ÑģÐ¿

-0.14

ISI

-0.13

Ð½Ð¾Ñģ

-0.13

 Infer

-0.13

 charity

-0.13

POSITIVE LOGITS

 Times

0.27

dna

0.25

 TIMES

0.25

Times

0.24

TO

0.21

DNA

0.21

DNA

0.20

 times

0.20

 Hind

0.20

PT

0.19

Activations Density 0.082%

references to news sources and article citations

No Comments

No Known Activations

references to news sources and article citations

No Comments

No Known Activations