INDEX

Explanations

references to "dark" themes, concepts, or entities

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

-0.17

pus

-0.15

ointed

-0.15

.scalablytyped

-0.15

aze

-0.15

_DC

-0.14

alue

-0.14

 é¦Ļ

-0.14

ekt

-0.14

aler

-0.14

POSITIVE LOGITS

ened

0.45

ening

0.38

ness

0.25

ling

0.23

room

0.23

ens

0.22

horse

0.21

eness

0.21

ly

0.21

 Ages

0.21

Activations Density 0.022%

references to "dark" themes, concepts, or entities

No Comments

No Known Activations