INDEX

Explanations

phrases that indicate logical connections or clarifications in arguments

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 22-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L22R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.22.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

oui

-0.15

 *)((

-0.15

viders

-0.14

ï¼ĮçĦ¶åĲİ

-0.14

leton

-0.14

Ú©Ø§Øª

-0.14

lette

-0.13

ahl

-0.13

ãĥ³ãĤº

-0.13

loquent

-0.13

POSITIVE LOGITS

 majority

0.30

 there

0.23

 till

0.22

few

0.22

 Majority

0.20

 apart

0.20

lot

0.19

 mostly

0.18

 lots

0.18

 during

0.17

Activations Density 0.367%

phrases that indicate logical connections or clarifications in arguments

No Comments

No Known Activations