INDEX

Explanations

instances of the word "nothing" and its variations, suggesting a focus on themes of lack or absence

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

 Thornton

-0.15

mont

-0.15

 something

-0.14

adiens

-0.14

ivel

-0.14

antar

-0.14

inel

-0.14

 ExecutionContext

-0.14

ME

-0.13

emmel

-0.13

POSITIVE LOGITS

 else

0.30

ness

0.28

else

0.23

/no

0.21

Else

0.21

wrong

0.19

 ELSE

0.19

_else

0.19

NESS

0.19

	else

0.19

Activations Density 0.033%

instances of the word "nothing" and its variations, suggesting a focus on themes of lack or absence

No Comments

No Known Activations