INDEX

Explanations

substrings or prefixes that are present in names or titles

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

elfast

-0.18

arness

-0.17

ourg

-0.17

MBOL

-0.15

ĥĿ

-0.15

ringe

-0.15

 onNext

-0.14

ãĥ¥ãĥ¼

-0.14

elper

-0.14

ceptar

-0.14

POSITIVE LOGITS

olini

0.15

ism

0.15

joy

0.15

ane

0.14

etta

0.14

cestor

0.14

 Lincoln

0.14

ure

0.14

jack

0.14

ÑĮÑı

0.14

Activations Density 0.026%

substrings or prefixes that are present in names or titles

No Comments

No Known Activations