INDEX

Explanations

phrases that indicate methods, instructions, or ways to accomplish tasks

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

 only

-0.15

 Favor

-0.14

See

-0.14

 whatever

-0.14

endar

-0.14

çľĭçľĭ

-0.14

dna

-0.14

 Need

-0.14

Don

-0.13

ando

-0.13

POSITIVE LOGITS

 best

0.34

best

0.30

 proceed

0.28

 Proceed

0.27

 BEST

0.25

-best

0.24

(best

0.24

Best

0.24

 approach

0.23

_best

0.23

Activations Density 0.089%

phrases that indicate methods, instructions, or ways to accomplish tasks

No Comments

No Known Activations