INDEX

Explanations

terms related to cultural or historical sites and landmarks

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

Ã¨

-0.18

 Gems

-0.15

oran

-0.15

him

-0.14

æ©

-0.14

ucene

-0.14

-0.13

esto

-0.13

oram

-0.13

POSITIVE LOGITS

IRA

0.18

erral

0.18

Ã³t

0.17

apot

0.17

avra

0.17

ossal

0.17

apon

0.16

ira

0.16

abras

0.16

usra

0.16

Activations Density 0.002%

terms related to cultural or historical sites and landmarks

No Comments

No Known Activations