INDEX

Explanations

references to events and their associated details, such as availability and specifications

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

 Gram

-0.15

/stat

-0.15

Ã¼lÃ¼k

-0.14

iyim

-0.14

 linh

-0.14

 stat

-0.13

pta

-0.13

 ì´

-0.13

 arada

-0.13

ROME

-0.13

POSITIVE LOGITS

 another

0.21

 similarly

0.21

another

0.20

 elsewhere

0.19

see

0.19

 other

0.19

åı¦ä¸Ģ

0.18

 separately

0.18

uche

0.17

see

0.17

Activations Density 0.314%

references to events and their associated details, such as availability and specifications

No Comments

No Known Activations