INDEX

Explanations

references to attention-grabbing actions or behaviors

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

unda

-0.16

ÃĹ</

-0.15

çľŁæŃ£

-0.14

zen

-0.14

irl

-0.14

uda

-0.13

zv

-0.13

 refresh

-0.13

rip

-0.13

 ÑģÐ¿ÑĢÐ°Ð²Ð¶

-0.13

POSITIVE LOGITS

 convenient

0.25

 Convenient

0.22

 convenience

0.20

 conveniently

0.19

 cheap

0.19

 Convenience

0.17

PR

0.17

 profitable

0.17

appe

0.16

 popularity

0.16

Activations Density 0.745%

references to attention-grabbing actions or behaviors

No Comments

No Known Activations

references to attention-grabbing actions or behaviors

No Comments

No Known Activations