INDEX

Explanations

questions beginning with "How."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

uÅŁ

-0.15

ksen

-0.14

ÙħÙĪÙĦ

-0.14

ÙĤØ¨

-0.13

Ø¯ÙĬØ«

-0.13

trusted

-0.13

artz

-0.13

olkien

-0.13

nier

-0.13

oster

-0.13

POSITIVE LOGITS

do

0.22

 does

0.21

 Does

0.19

did

0.18

Does

0.17

to

0.16

ells

0.16

ine

0.15

does

0.15

rah

0.15

Activations Density 0.040%

questions beginning with "How."

No Comments

No Known Activations