INDEX

Explanations

phrases that convey documentation or reporting actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 22-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L22R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.22.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

Tho

-0.14

.enterprise

-0.13

eld

-0.13

alnÄ±z

-0.13

ilio

-0.13

olumn

-0.13

åĥķ

-0.12

sto

-0.12

oader

-0.12

 indifferent

-0.12

POSITIVE LOGITS

croft

0.15

 Bros

0.14

ricks

0.14

onga

0.14

 Kimber

0.13

istrovstvÃŃ

0.13

±Ð¾ÑĤ

0.13

Ã¡zev

0.13

åĵ¡

0.13

 Preis

0.12

Activations Density 0.199%

phrases that convey documentation or reporting actions

No Comments

No Known Activations