INDEX

Explanations

conditional or advisory phrases suggesting actions or options

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

 Screw

-0.14

Ð¾Ð»Ð¾ÑĤ

-0.14

erm

-0.14

ioneer

-0.13

.invoke

-0.13

esti

-0.13

dia

-0.13

èĿ

-0.13

æĿĳ

-0.13

abella

-0.13

POSITIVE LOGITS

łéĻ¤

0.14

hardt

0.13

angep

0.13

_quota

0.13

blick

0.12

Ã¡nh

0.12

PointerType

0.12

*this

0.12

chron

0.12

Activations Density 0.090%

conditional or advisory phrases suggesting actions or options

No Comments

No Known Activations