INDEX

Explanations

structured data or tabular information

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

]âĢı

-0.15

adb

-0.14

uft

-0.14

_STRUCTURE

-0.13

Twenty

-0.13

 Twenty

-0.13

osing

-0.13

SubMenu

-0.13

redo

-0.13

POSITIVE LOGITS

0.17

etc

0.14

inf

0.14

apat

0.14

uz

0.14

ÙĨØ§ÙĨ

0.13

 Corpus

0.13

ÑĢÑĥÐ¿

0.13

ding

0.13

Activations Density 0.012%

structured data or tabular information

No Comments

No Known Activations