INDEX

Explanations

conditional statements and phrases indicating possibilities or options

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

 Anything

-0.14

het

-0.14

Ð»ÑİÐ±

-0.14

YLON

-0.14

anything

-0.14

nik

-0.13

zen

-0.13

Anything

-0.13

ãģ¨ãģĵãĤį

-0.13

ë³µ

-0.13

POSITIVE LOGITS

 none

0.27

 neither

0.24

 NONE

0.21

 None

0.20

 budget

0.19

 budgets

0.19

None

0.18

 money

0.18

none

0.17

 subt

0.17

Activations Density 0.092%

conditional statements and phrases indicating possibilities or options

No Comments

No Known Activations