INDEX

Explanations

elements related to programming questions and technical issues

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 27-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L27R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.27.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

.fhir

-0.16

Alright

-0.16

Amt

-0.16

 hence

-0.15

 indiv

-0.14

loggedIn

-0.14

 seper

-0.14

 amongst

-0.14

 seperate

-0.14

Similarly

-0.14

POSITIVE LOGITS

 beside

0.18

 theirs

0.17

 maybe

0.17

 comport

0.16

 politic

0.16

 Ä°t

0.15

 like

0.15

 especial

0.15

 protections

0.15

 defer

0.14

Activations Density 1.360%

elements related to programming questions and technical issues

No Comments

No Known Activations