INDEX

Explanations

references to loss or absence

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

ÑĥÑĢÐ½

-0.18

esc

-0.17

arge

-0.15

tÄĽ

-0.14

cole

-0.14

 ÙħÙĪ

-0.14

Å¡k

-0.14

tog

-0.13

PKG

-0.13

POSITIVE LOGITS

 Bars

0.23

Bars

0.21

 Strategy

0.20

 symbol

0.20

 strategy

0.20

 bars

0.19

 Symbol

0.19

(strategy

0.18

-bars

0.17

bars

0.17

Activations Density 0.000%

references to loss or absence

No Comments

No Known Activations