INDEX

Explanations

phrases indicating progress and completion of tasks or events

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

 never

-0.20

 immediately

-0.19

 still

-0.18

still

-0.18

 sofort

-0.18

 masih

-0.18

 stayed

-0.18

 NEVER

-0.17

remain

-0.17

 quickly

-0.17

POSITIVE LOGITS

 fully

0.28

 finishes

0.27

 finish

0.27

 stabil

0.23

 finished

0.22

 finishing

0.21

 figure

0.21

 complete

0.20

 Fully

0.20

 hopefully

0.20

Activations Density 0.283%

phrases indicating progress and completion of tasks or events

No Comments

No Known Activations