INDEX

Explanations

references to concern or abnormality in situations

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

mps

-0.16

ramer

-0.15

fait

-0.15

éŀ

-0.14

 Wyatt

-0.14

èįĴ

-0.14

ÙĪÙĦÙĪØ¬

-0.13

 khÃ³

-0.13

oyal

-0.13

çī¹èī²

-0.13

POSITIVE LOGITS

 wrong

0.40

wrong

0.35

 Wrong

0.33

Wrong

0.29

 WRONG

0.29

_wrong

0.24

 fish

0.23

fish

0.19

 Fish

0.19

 wrongful

0.18

Activations Density 0.060%

references to concern or abnormality in situations

No Comments

No Known Activations