INDEX

Explanations

references to titles, particularly related to media or literature

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

 HÃ¼s

-0.16

ä¼

-0.16

oldem

-0.16

Ð°ÑĢÑĩ

-0.14

uzey

-0.14

vero

-0.14

roe

-0.14

[axis

-0.14

umas

-0.14

scl

-0.14

POSITIVE LOGITS

ervers

0.16

 Bowen

0.15

Capability

0.15

endez

0.15

leta

0.14

 Chief

0.14

é¦ĸ

0.14

<decltype

0.13

 Away

0.13

Obs

0.13

Activations Density 0.003%

references to titles, particularly related to media or literature

No Comments

No Known Activations