INDEX

Explanations

references to rankings or numerical standings

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 27-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L27R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.27.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

nya

-0.17

proof

-0.17

nech

-0.15

 Ø¢Ø®Ø±

-0.15

otto

-0.14

lop

-0.14

nee

-0.14

ppard

-0.14

uku

-0.14

POSITIVE LOGITS

xious

0.22

ises

0.19

things

0.19

xon

0.19

oit

0.18

veau

0.17

zzle

0.17

 doubt

0.17

venta

0.17

strand

0.16

Activations Density 0.008%

references to rankings or numerical standings

No Comments

No Known Activations