INDEX

Explanations

quality descriptors and evaluations of products or services

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

 coolest

-0.18

ngÃ¶r

-0.17

ampus

-0.16

hest

-0.16

 cute

-0.16

 interess

-0.15

ander

-0.15

ribly

-0.15

 Nice

-0.15

 Cute

-0.15

POSITIVE LOGITS

top

0.44

 outstanding

0.42

 excellent

0.41

 exceptional

0.35

 Outstanding

0.34

 Excellent

0.32

top

0.32

 stellar

0.31

Excellent

0.30

 fault

0.30

Activations Density 0.324%

quality descriptors and evaluations of products or services

No Comments

No Known Activations