INDEX

Explanations

complex sentence structures and phrases related to ethical dilemmas

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

ï»¿#

-0.15

 láº«n

-0.14

akest

-0.14

ãģłãģĳãģ§

-0.14

prise

-0.14

åłĤ

-0.13

ollapse

-0.12

ãĤĦãĤĭ

-0.12

TestingModule

-0.12

emain

-0.12

POSITIVE LOGITS

 maybe

1.16

 perhaps

1.15

 possibly

1.02

maybe

0.98

perhaps

0.94

 Maybe

0.88

 Perhaps

0.84

Maybe

0.84

 vielleicht

0.82

possibly

0.80

Activations Density 1.340%

complex sentence structures and phrases related to ethical dilemmas

No Comments

No Known Activations