INDEX

Explanations

phrases and sentiments associated with moral judgments and emotional responses

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

ear

-0.15

zm

-0.15

 culpa

-0.14

 invo

-0.14

 Stock

-0.13

SCI

-0.13

ecek

-0.13

ough

-0.13

POSITIVE LOGITS

gew

0.15

à¸§à¸Ķ

0.15

ycastle

0.15

insk

0.14

duto

0.14

ÎºÎ¿Ïį

0.14

Î¸Îµ

0.14

dol

0.14

éĢļ

0.13

geois

0.13

Activations Density 0.168%

phrases and sentiments associated with moral judgments and emotional responses

No Comments

No Known Activations