INDEX

Explanations

connections and interactions between two distinct entities or concepts

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

ry

-0.15

thon

-0.15

jak

-0.14

rules

-0.14

 rules

-0.14

 Rules

-0.14

ulan

-0.14

agens

-0.14

uj

-0.14

POSITIVE LOGITS

.gwt

0.16

 complementary

0.15

WX

0.15

ä¹ĭéĹ´

0.15

 ëĳĲ

0.14

-two

0.14

ionic

0.14

 Ð´Ð²Ñĸ

0.14

two

0.14

 sides

0.14

Activations Density 0.313%

connections and interactions between two distinct entities or concepts

No Comments

No Known Activations