INDEX

Explanations

terms related to feet and foot positioning

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 16-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L16R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.16.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

fy

-0.17

ittal

-0.16

rtl

-0.15

ĸ

-0.14

ustomer

-0.14

ming

-0.14

hazi

-0.14

Bul

-0.14

hit

-0.14

hwnd

-0.14

POSITIVE LOGITS

prints

0.21

 Foot

0.21

-foot

0.20

sie

0.20

 foot

0.20

Foot

0.20

foot

0.19

step

0.18

bridge

0.18

 footprint

0.17

Activations Density 0.029%

terms related to feet and foot positioning

No Comments

No Known Activations