INDEX

Explanations

instances of surprise and conversational exchanges

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

affen

-0.17

ipop

-0.15

illard

-0.15

agon

-0.15

imi

-0.15

urum

-0.15

ÄĽt

-0.14

Ø¹ÙĦÙĪÙħØ§Øª

-0.14

usters

-0.14

.removeFrom

-0.14

POSITIVE LOGITS

 notice

0.42

see

0.39

 notices

0.38

 sees

0.38

 noticed

0.35

 noticing

0.34

notice

0.32

 seeing

0.32

 Notice

0.31

see

0.31

Activations Density 0.139%

instances of surprise and conversational exchanges

No Comments

No Known Activations