INDEX

Explanations

references to educational or instructive processes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

ofil

-0.14

ixed

-0.13

↵

-0.13

avin

-0.13

çĽ¸

-0.13

ERING

-0.13

ä¸

-0.13

kaar

-0.13

POSITIVE LOGITS

eyond

0.26

 beyond

0.25

Beyond

0.22

 deeper

0.21

 larger

0.21

 Beyond

0.21

 wider

0.20

 Larger

0.20

expanded

0.19

 bigger

0.19

Activations Density 0.011%

references to educational or instructive processes

No Comments

No Known Activations