INDEX

Explanations

titles and references to literary or artistic works

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 23-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L23R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.23.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

vision

-0.18

vote

-0.18

agas

-0.15

ument

-0.15

vendors

-0.15

vehicles

-0.15

votes

-0.14

vendor

-0.14

Vision

-0.14

variant

-0.14

POSITIVE LOGITS

GV

0.27

PV

0.23

SVC

0.23

GV

0.23

(V

0.23

pv

0.22

MV

0.21

PV

0.21

VV

0.20

 metav

0.19

Activations Density 0.168%

titles and references to literary or artistic works

No Comments

No Known Activations