INDEX

Explanations

terms related to various professional and creative industries

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

Negative Logits

zano

-0.17

ieren

-0.16

lich

-0.16

oir

-0.15

odor

-0.15

quez

-0.14

vro

-0.14

 VÅ¡

-0.14

utr

-0.14

arih

-0.14

POSITIVE LOGITS

 itself

0.25

ÐŁÐļ

0.17

osphere

0.16

åľĪ

0.16

 writ

0.16

-wide

0.15

DRV

0.15

 landscape

0.15

MMC

0.15

/world

0.14

Activations Density 0.156%

terms related to various professional and creative industries

No Known Activations