INDEX

Explanations

interrogative words or phrases related to identification or selection

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 16-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L16R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.16.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

uto

-0.17

tran

-0.16

egg

-0.15

EGA

-0.15

ming

-0.14

upe

-0.14

Ð¾ÑģÑĮ

-0.14

 tÃ¢m

-0.13

uros

-0.13

uese

-0.13

POSITIVE LOGITS

soever

0.25

 kind

0.19

/how

0.17

 Ð¸Ð¼ÐµÐ½Ð½Ð¾

0.16

 direction

0.16

 parts

0.16

 kinds

0.15

 Wich

0.15

-sex

0.15

erry

0.15

Activations Density 0.032%

interrogative words or phrases related to identification or selection

No Comments

No Known Activations