INDEX

Explanations

direct references to the word "you" in various contexts

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

 Uncomment

-0.15

rawer

-0.15

usher

-0.15

ogui

-0.14

yny

-0.14

 Ð²Ð´ÑĢÑĥÐ³

-0.14

onom

-0.14

asser

-0.14

 numberWith

-0.14

arp

-0.14

POSITIVE LOGITS

 forgot

0.19

 said

0.19

 seem

0.19

 seems

0.17

 mention

0.17

 mentioned

0.17

 mileage

0.17

 seemed

0.16

å¿ĺ

0.16

Activations Density 0.044%

direct references to the word "you" in various contexts

No Comments

No Known Activations