INDEX

Explanations

references to image sources and credits

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

esseract

-0.17

egend

-0.16

.writeValue

-0.16

ildo

-0.15

åľ¨çº¿éĺħè¯»

-0.15

ÎºÏģÎ±

-0.15

ummer

-0.15

usra

-0.15

ilyn

-0.15

 Redistributions

-0.15

POSITIVE LOGITS

 stock

0.58

stock

0.54

 Stock

0.53

Stock

0.51

-stock

0.46

 STOCK

0.45

_stock

0.39

.stock

0.39

 royalty

0.37

(stock

0.37

Activations Density 0.075%

references to image sources and credits

No Comments

No Known Activations