INDEX

Explanations

phrases that indicate progression or movement through a narrative or experience

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

kova

-0.18

ushima

-0.17

eland

-0.16

acular

-0.16

Ð»Ð¾Ð¹

-0.15

ãĥ¥

-0.15

bles

-0.15

ëģĶ

-0.14

ailles

-0.14

 láº¡i

-0.14

POSITIVE LOGITS

side

0.38

-side

0.30

SIDE

0.27

shore

0.24

 side

0.22

sid

0.22

-term

0.21

0.20

/about

0.20

ue

0.19

Activations Density 0.029%

phrases that indicate progression or movement through a narrative or experience

No Comments

No Known Activations

phrases that indicate progression or movement through a narrative or experience

No Comments

No Known Activations