INDEX

Explanations

boolean values, particularly those related to state changes or conditions in a programming context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 10-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L10R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.10.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

MMI

-0.15

Raq

-0.14

ÙĲÙĩ

-0.14

chemas

-0.14

 ÎłÏģÏī

-0.14

ogs

-0.14

 ê³µ

-0.13

Ø±ÛĮÙĤ

-0.13

ĮĴ

-0.13

abin

-0.13

POSITIVE LOGITS

ero

0.19

ERO

0.17

eum

0.16

 certificates

0.15

oped

0.15

atin

0.15

ihn

0.14

 Bond

0.14

 Steel

0.14

 wink

0.14

Activations Density 0.179%

boolean values, particularly those related to state changes or conditions in a programming context

No Comments

No Known Activations