INDEX

Explanations

expressions indicating hesitation or reluctance

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

 beyond

-0.17

 Beyond

-0.17

enz

-0.16

 Wand

-0.16

illi

-0.14

 almost

-0.14

ermalink

-0.14

ocused

-0.14

ÄĽti

-0.14

epar

-0.13

POSITIVE LOGITS

nor

0.59

Nor

0.43

nor

0.43

Nor

0.41

 anymore

0.34

NOR

0.33

 either

0.30

 EITHER

0.27

either

0.26

 Either

0.25

Activations Density 0.298%

expressions indicating hesitation or reluctance

No Comments

No Known Activations