INDEX

Explanations

references to concepts of identity and societal roles

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

anca

-0.15

ommen

-0.15

UF

-0.14

bare

-0.14

opers

-0.14

PEC

-0.14

ocre

-0.14

Łèĥ½

-0.14

otal

-0.14

 barely

-0.14

POSITIVE LOGITS

 automatically

0.37

 automatic

0.28

 Automatically

0.28

autom

0.26

 necessarily

0.25

 defaulted

0.24

 inevitable

0.23

 inevitably

0.23

 automÃ¡ticamente

0.23

 naturally

0.22

Activations Density 0.049%

references to concepts of identity and societal roles

No Comments

No Known Activations