INDEX

Explanations

references to conferences and their proceedings

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

ling

-0.16

arna

-0.16

ush

-0.16

ifestyles

-0.15

ypad

-0.15

are

-0.15

rap

-0.15

opro

-0.15

la

-0.15

fully

-0.14

POSITIVE LOGITS

erialize

0.17

ãĥ³ãĤ¹

0.16

estro

0.16

cci

0.15

zimmer

0.15

GenerationStrategy

0.14

wahl

0.14

OUNCE

0.14

WindowTitle

0.14

azzi

0.14

Activations Density 0.017%

references to conferences and their proceedings

No Comments

No Known Activations