INDEX

Explanations

references to different sections or parts of a larger work or series

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 20-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L20R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.20.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

eer

-0.17

ameleon

-0.14

ocode

-0.14

è¼

-0.14

ToFit

-0.14

Ĥæķ°

-0.14

 benim

-0.14

ilent

-0.14

arrison

-0.13

et

-0.13

POSITIVE LOGITS

 Part

0.22

 part

0.21

II

0.20

III

0.19

 PART

0.18

Part

0.17

II

0.17

-part

0.17

part

0.16

PART

0.16

Activations Density 0.047%

references to different sections or parts of a larger work or series

No Comments

No Known Activations