INDEX

Explanations

function definitions and method signatures in code

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

anian

-0.18

ª

-0.14

non

-0.14

alim

-0.14

!,↵

-0.14

IU

-0.13

tml

-0.13

 chip

-0.13

beit

-0.13

illery

-0.13

POSITIVE LOGITS

 nothing

0.50

nothing

0.41

 NOTHING

0.41

 Nothing

0.40

Nothing

0.36

 nada

0.31

 nichts

0.31

 rien

0.29

 Ð½Ð¸ÑĩÐµÐ³Ð¾

0.29

){}↵

0.25

Activations Density 0.106%

function definitions and method signatures in code

No Comments

No Known Activations