INDEX

Explanations

references to anger and related emotions

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

ksen

-0.16

itive

-0.16

atives

-0.15

etto

-0.15

podob

-0.15

ardon

-0.14

tracer

-0.14

 Garn

-0.14

etten

-0.14

raya

-0.14

POSITIVE LOGITS

Ang

0.27

els

0.22

Ang

0.22

gota

0.21

Ð»Ð¸Ð¹

0.20

ELS

0.20

los

0.20

eline

0.19

(ang

0.19

ang

0.19

Activations Density 0.014%

references to anger and related emotions

No Comments

No Known Activations