INDEX

Explanations

references to promises and pledges made by individuals, particularly in a political context

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

omm

-0.16

erland

-0.16

elor

-0.15

ktop

-0.15

akis

-0.15

ENAME

-0.14

elah

-0.14

ÑĢÐ¾Ð½

-0.14

Ã¡zev

-0.14

POSITIVE LOGITS

 promises

0.24

 promise

0.22

pled

0.18

 pledge

0.18

.promise

0.17

promise

0.17

 pledges

0.17

/prom

0.16

 promised

0.16

vow

0.16

Activations Density 0.142%

references to promises and pledges made by individuals, particularly in a political context

No Comments

No Known Activations