INDEX

Explanations

words related to securing or enclosing objects

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L1R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.1.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

lette

-0.16

ike

-0.16

ureka

-0.15

Ing

-0.15

 æĮ

-0.15

uges

-0.14

 recess

-0.14

alogy

-0.14

 Spear

-0.14

Strict

-0.14

POSITIVE LOGITS

ively

0.18

appoint

0.17

pole

0.17

ables

0.15

robe

0.15

abile

0.15

ability

0.15

 doorstep

0.15

 indem

0.15

gate

0.14

Activations Density 0.005%

words related to securing or enclosing objects

No Comments

No Known Activations