INDEX

Explanations

terms related to performance and quality in the context of services and products

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

--

-0.16

-turned

-0.14

-related

-0.13

unch

-0.13

 aforementioned

-0.13

avian

-0.13

-/

-0.13

-style

-0.12

/"

-0.12

 (“

-0.12

POSITIVE LOGITS

Lets

0.14

GuidId

0.13

 rouge

0.13

iliz

0.13

BOSE

0.13

ookies

0.12

HeaderCode

0.12

Nuitka

0.12

_Parms

0.12

adnÃŃ

0.12

Activations Density 0.001%

terms related to performance and quality in the context of services and products

No Comments

No Known Activations