INDEX

Explanations

references to mathematical or logical operations

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

 Gates

-0.15

odium

-0.15

Lug

-0.15

ervas

-0.14

ÅĽ

-0.14

utdown

-0.14

_HAL

-0.14

iform

-0.14

onom

-0.14

_HEAP

-0.14

POSITIVE LOGITS

yz

0.16

iero

0.15

abee

0.15

UBLIC

0.15

ENE

0.15

Ð³Ð»ÑıÐ´

0.14

ÙĪÛĮÙĩ

0.14

Ã±ana

0.14

 shar

0.14

 Nack

0.14

Activations Density 0.120%

references to mathematical or logical operations

No Comments

No Known Activations