INDEX

Explanations

programming constructs related to classes and templates

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

jez

-0.15

igne

-0.15

enger

-0.15

fare

-0.15

ivery

-0.14

 Hayward

-0.14

lore

-0.14

Tay

-0.14

Ign

-0.14

engers

-0.14

POSITIVE LOGITS

itk

0.20

 Dice

0.18

.nii

0.17

PerPixel

0.16

 ÙĨÙģ

0.16

 Image

0.15

DIC

0.15

PET

0.15

InputLabel

0.15

Activations Density 0.009%

programming constructs related to classes and templates

No Comments

No Known Activations