INDEX

Explanations

specific measurements and numerical values related to instructions

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

bara

-0.16

odia

-0.16

lac

-0.14

_overflow

-0.14

 under

-0.14

 deformation

-0.14

-schema

-0.14

ãģľ

-0.14

Ð»Ð°Ñĩ

-0.14

vvm

-0.13

POSITIVE LOGITS

_Template

0.18

 Incontri

0.17

 templates

0.17

 à¤®à¤¦

0.17

TemplateName

0.17

 template

0.17

 Template

0.16

-template

0.15

 TEMPLATE

0.15

template

0.15

Activations Density 0.037%

specific measurements and numerical values related to instructions

No Known Activations