INDEX

Explanations

references to art-related entities and events

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

wit

-0.14

 McCorm

-0.13

ØªØ§

-0.13

acic

-0.13

gii

-0.13

Ì£

-0.13

engkap

-0.13

 universities

-0.13

tas

-0.13

ignon

-0.13

POSITIVE LOGITS

 Centre

0.26

 Company

0.24

 Center

0.24

works

0.21

Center

0.21

Works

0.20

Centre

0.20

æĹıèĩªæ²»

0.20

center

0.20

centre

0.20

Activations Density 0.326%

references to art-related entities and events

No Comments

No Known Activations