INDEX

Explanations

terms related to default settings or values in configurations

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

à¤łà¤¨

-0.16

IRC

-0.16

erman

-0.16

ibar

-0.15

eka

-0.15

à¸£

-0.14

outer

-0.14

 permit

-0.14

/he

-0.14

essel

-0.14

POSITIVE LOGITS

/default

0.23

=default

0.19

ê°Ĵ

0.17

/native

0.16

cott

0.16

.Formatter

0.16

ies

0.16

ONGL

0.16

/base

0.16

ÑĪÑĤ

0.15

Activations Density 0.030%

terms related to default settings or values in configurations

No Comments

No Known Activations