INDEX

Explanations

instances of items or topics that are described with elaborate details, particularly focusing on the use of "with" in descriptive phrases

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 6-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L6R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.6.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

odon

-0.18

ignum

-0.16

à¸±à¸ķà¸ĸ

-0.15

.Glide

-0.14

okane

-0.14

tep

-0.14

odge

-0.14

'&'

-0.14

asca

-0.14

ange

-0.13

POSITIVE LOGITS

INU

0.19

 EntryPoint

0.14

RS

0.13

inu

0.13

ente

0.13

entr

0.13

forme

0.13

kr

0.13

UTH

0.13

Activations Density 0.099%

instances of items or topics that are described with elaborate details, particularly focusing on the use of "with" in descriptive phrases

No Comments

No Known Activations