INDEX

Explanations

references to legal or formal identifiers

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 6-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L6R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.6.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

chl

-0.16

cale

-0.16

aic

-0.15

ua

-0.15

oin

-0.14

enty

-0.14

zw

-0.14

uche

-0.13

 Byrne

-0.13

POSITIVE LOGITS

laz

0.19

	Copyright

0.16

ìĤ¬ìĿ´

0.16

ukan

0.15

Î¼Î¼

0.15

 ZÃ¡p

0.14

å¡ļ

0.14

'{@

0.14

uteur

0.14

Affected

0.14

Activations Density 0.019%

references to legal or formal identifiers

No Comments

No Known Activations