INDEX

Explanations

descriptions of visual clarity and quality

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

 tension

-0.16

idian

-0.16

.skin

-0.15

 compat

-0.15

ilon

-0.15

oppins

-0.15

 scent

-0.14

ge

-0.14

 scen

-0.14

oe

-0.14

POSITIVE LOGITS

 fuzz

0.31

 distortion

0.29

blur

0.29

 fuzzy

0.29

 cloud

0.27

fog

0.27

 blur

0.26

 distorted

0.26

cloud

0.25

Blur

0.25

Activations Density 0.177%

descriptions of visual clarity and quality

No Comments

No Known Activations