INDEX

Explanations

instances of significant numbers or identifiers indicating quantities or measures

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

 oslo

-0.15

uggy

-0.15

ali

-0.14

entiful

-0.14

ÑĨÑĸÐ¹Ð½Ð°

-0.14

ajo

-0.14

alytics

-0.14

ugged

-0.14

antis

-0.14

agg

-0.13

POSITIVE LOGITS

/&

0.16

cctor

0.15

StreamReader

0.15

inth

0.15

 amateur

0.15

łĢ

0.14

WND

0.14

 outside

0.14

urette

0.14

]=>

0.14

Activations Density 0.018%

instances of significant numbers or identifiers indicating quantities or measures

No Comments

No Known Activations