INDEX

Explanations

instances of numerical data and annotations related to timing or identifiers

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

/datatables

-0.15

aal

-0.15

utzer

-0.14

 Neville

-0.14

.Magenta

-0.14

turnstile

-0.14

phia

-0.14

icari

-0.13

ynom

-0.13

 Modal

-0.13

POSITIVE LOGITS

 orient

0.30

 controls

0.26

 Controls

0.25

 control

0.24

Controls

0.22

Ori

0.22

control

0.21

_controls

0.21

.control

0.21

 Control

0.21

Activations Density 0.001%

instances of numerical data and annotations related to timing or identifiers

No Comments

No Known Activations