INDEX

Explanations

expressions of value and worth in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

ober

-0.16

esium

-0.15

bz

-0.15

ooth

-0.14

 factual

-0.14

ÑĥÑħ

-0.14

ked

-0.14

Ľi

-0.14

 stan

-0.14

 embarrass

-0.14

POSITIVE LOGITS

 worth

0.41

 Worth

0.40

worth

0.35

 outweigh

0.27

 worthwhile

0.26

 justify

0.23

 Reward

0.23

 reward

0.22

 justified

0.22

 rewards

0.21

Activations Density 0.155%

expressions of value and worth in various contexts

No Comments

No Known Activations