INDEX

Explanations

structured sections and headings in a document

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

urator

-0.16

idad

-0.14

 GENERIC

-0.14

atab

-0.14

mez

-0.14

glich

-0.14

_UNUSED

-0.13

 creature

-0.13

 fortunes

-0.13

_UNICODE

-0.13

POSITIVE LOGITS

}↵↵

0.17

Ð°Ð½Ð¸ÑĨ

0.15

èĻİ

0.15

}\

0.14

 section

0.14

}↵

0.14

âĺĨ

0.14

oose

0.14

/background

0.14

osl

0.14

Activations Density 0.016%

structured sections and headings in a document

No Comments

No Known Activations