INDEX

Explanations

discussions about personal achievements and accomplishments

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 annual

-0.17

-0.15

 Annual

-0.15

apus

-0.15

etin

-0.15

 Grass

-0.14

hashed

-0.14

 pozdÄĽ

-0.14

 enim

-0.14

 underwent

-0.13

POSITIVE LOGITS

 bisher

0.48

 zatÃŃm

0.38

 till

0.37

 already

0.37

çĽ®åīį

0.36

 Already

0.34

already

0.34

 dosud

0.33

Already

0.33

yet

0.32

Activations Density 0.349%

discussions about personal achievements and accomplishments

No Comments

No Known Activations