INDEX

Explanations

references to structured divisions or groupings within content

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

endar

-0.15

 repertoire

-0.14

 arena

-0.14

USTER

-0.14

sit

-0.14

 situation

-0.14

 Seite

-0.14

oney

-0.14

alytics

-0.14

jt

-0.14

POSITIVE LOGITS

 sections

0.30

 zones

0.28

 stages

0.28

 segments

0.26

 parts

0.26

 tiers

0.26

 modules

0.26

 distinct

0.25

 subsection

0.25

 divisions

0.24

Activations Density 0.332%

references to structured divisions or groupings within content

No Comments

No Known Activations