INDEX

Explanations

phrases indicating quantity or amount

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

ovsky

-0.17

soever

-0.17

xic

-0.16

æĽ´å¤ļ

-0.16

 Some

-0.15

 slightest

-0.14

 some

-0.14

 quelques

-0.14

ãĤĪãģĨãģª

-0.14

 zbyt

-0.14

POSITIVE LOGITS

 different

0.22

 them

0.20

different

0.19

 effort

0.18

us

0.18

 money

0.17

 emphasis

0.17

ones

0.17

them

0.17

/all

0.17

Activations Density 0.068%

phrases indicating quantity or amount

No Comments

No Known Activations