INDEX

Explanations

words related to programming languages and frameworks, especially those associated with the Sass preprocessor

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

onder

-0.17

oster

-0.15

allery

-0.15

GDK

-0.15

asaki

-0.15

aise

-0.15

ISIBLE

-0.14

 agre

-0.14

 Template

-0.14

umont

-0.14

POSITIVE LOGITS

\OptionsResolver

0.15

 progressive

0.15

uristic

0.15

 Volt

0.15

Kok

0.15

kok

0.14

 bust

0.14

BLOCK

0.14

elm

0.14

ugg

0.14

Activations Density 0.020%

words related to programming languages and frameworks, especially those associated with the Sass preprocessor

No Comments

No Known Activations