INDEX

Explanations

quantitative data and statistical references

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 28-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L28R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.28.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

ÑĢÐ°Ð²Ð¸

-0.15

illet

-0.15

ÐµÐ»Ð¸Ñĩ

-0.15

arah

-0.14

Ð¸Ð½Ñĥ

-0.14

ouv

-0.14

.synthetic

-0.14

åħ¶ä¸Ń

-0.14

_TEXTURE

-0.14

POSITIVE LOGITS

 therefore

0.21

 because

0.20

 pois

0.19

po

0.19

 pues

0.19

mal

0.18

 puesto

0.18

 porque

0.18

ni

0.17

åĽłä¸º

0.17

Activations Density 0.092%

quantitative data and statistical references

No Comments

No Known Activations