INDEX

Explanations

references to enjoyable experiences and positive activities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

alink

-0.14

igu

-0.14

 lÃ´

-0.14

cliffe

-0.14

 Scalars

-0.14

.sax

-0.13

redo

-0.13

aurus

-0.13

vester

-0.13

 Trad

-0.13

POSITIVE LOGITS

 overall

0.17

 thus

0.16

 these

0.14

 foregoing

0.14

overall

0.14

 morph

0.14

 bottom

0.13

 tonight

0.13

dorf

0.13

boarding

0.13

Activations Density 0.105%

references to enjoyable experiences and positive activities

No Comments

No Known Activations