INDEX

Explanations

discussions about resource allocation and efficiency

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 13-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L13R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.13.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

&_

-0.14

raries

-0.14

cket

-0.14

 only

-0.14

idon

-0.14

_soft

-0.14

zaw

-0.14

phin

-0.14

 threshold

-0.14

POSITIVE LOGITS

 instead

0.25

instead

0.24

 Instead

0.24

Instead

0.24

 Ð²Ð¼ÐµÑģÑĤ

0.18

 yerine

0.17

 tiáº¿t

0.16

 elsewhere

0.16

uis

0.16

-loader

0.15

Activations Density 0.184%

discussions about resource allocation and efficiency

No Comments

No Known Activations