INDEX

Explanations

references to frustration or difficulty in achieving tasks

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

uld

-0.17

ekim

-0.17

ubo

-0.15

idor

-0.15

 period

-0.14

peu

-0.14

ame

-0.14

ome

-0.14

éľĩ

-0.14

 Ð¿ÐµÑĢÐ¸Ð¾Ð´

-0.13

POSITIVE LOGITS

 forever

0.49

 ages

0.49

ages

0.39

 Ages

0.38

 FORE

0.37

 Forever

0.37

Forever

0.34

AGES

0.33

fore

0.31

age

0.28

Activations Density 0.097%

references to frustration or difficulty in achieving tasks

No Comments

No Known Activations