INDEX

Explanations

references to a specific designer or brand, particularly related to fashion

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

er

-0.23

erland

-0.18

erken

-0.18

isans

-0.17

erre

-0.17

erif

-0.17

ata

-0.17

eri

-0.16

erin

-0.16

ermo

-0.15

POSITIVE LOGITS

0.21

ond

0.20

es

0.18

ÄĽlÃŃ

0.17

ÅĻej

0.17

ador

0.16

orf

0.16

uras

0.16

ye

0.16

Activations Density 0.013%

references to a specific designer or brand, particularly related to fashion

No Comments

No Known Activations