INDEX

Explanations

references to authority and its impact on actions and outcomes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 31-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.31.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

 millenn

-0.16

 dramatic

-0.15

 erotiske

-0.15

alon

-0.14

 dramatically

-0.14

 mythology

-0.14

 erotik

-0.13

 ÑģÐ»Ð¸Ð²

-0.13

 drama

-0.13

Invoker

-0.13

POSITIVE LOGITS

0.24

[=

0.20

 aforementioned

0.18

 yourselves

0.17

 everyone

0.17

 everybody

0.17

Our

0.16

 very

0.16

 (...)

0.16

 Your

0.16

Activations Density 0.010%

references to authority and its impact on actions and outcomes

No Comments

No Known Activations