INDEX

Explanations

terms related to design, particularly in the context of jewelry and adornments

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 15-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L15R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.15.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

bar

-0.13

all

-0.13

Ant

-0.12

ather

-0.12

 Monkey

-0.12

lick

-0.12

roe

-0.12

istrovstvÃŃ

-0.12

 anth

-0.12

 Brett

-0.12

POSITIVE LOGITS

ationally

0.13

adors

0.13

scription

0.13

ATO

0.13

uge

0.13

OURS

0.13

:UITableView

0.13

NDER

0.13

Ãªte

0.12

_requested

0.12

Activations Density 0.265%

terms related to design, particularly in the context of jewelry and adornments

No Comments

No Known Activations