INDEX

Explanations

phrases or terms related to specific brands, models, or technical jargon within different contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

ÛĮÙĨÙĩ

-0.15

ieres

-0.14

esser

-0.14

arkin

-0.13

Ìģc

-0.13

flen

-0.12

Multiply

-0.12

mony

-0.12

----------------------------------------------------------------------

-0.12

rawl

-0.12

POSITIVE LOGITS

.scalablytyped

0.19

ALCHEMY

0.15

ì¹´ëĿ¼

0.13

umhur

0.13

":[{↵

0.13

osyal

0.13

 Dhabi

0.13

ESTAMP

0.13

utsche

0.13

/gtest

0.13

Activations Density 0.171%

phrases or terms related to specific brands, models, or technical jargon within different contexts

No Comments

No Known Activations