INDEX

Explanations

references to individuals taking action or needing to act

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 10-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L10R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.10.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

zin

-0.17

lej

-0.17

endoza

-0.15

culo

-0.15

à¤ĺ

-0.15

rome

-0.15

ĥģ

-0.15

ì¹´ëĿ¼

-0.15

spa

-0.14

ÄĻd

-0.14

POSITIVE LOGITS

 else

0.19

oni

0.15

 Schmidt

0.14

 crack

0.14

 somewhere

0.14

(s

0.14

 Somebody

0.14

ze

0.14

 call

0.13

 whom

0.13

Activations Density 0.077%

references to individuals taking action or needing to act

No Comments

No Known Activations