INDEX

Explanations

instances of the word "conducted" and its variations, indicating research activities or studies being carried out

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 2-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L2R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.2.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

 ç¸

-0.17

Binder

-0.16

ermo

-0.15

 GÃľ

-0.15

olut

-0.14

/t

-0.14

umen

-0.14

 compulsory

-0.14

undle

-0.14

avl

-0.14

POSITIVE LOGITS

 Zust

0.17

 Terminal

0.15

é¬

0.14

uess

0.14

 enqu

0.14

.normalized

0.14

COPE

0.13

stdarg

0.13

adj

0.13

Activations Density 0.020%

instances of the word "conducted" and its variations, indicating research activities or studies being carried out

No Comments

No Known Activations