INDEX

Explanations

instances of overflow or breach leading to physical consequences

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

ismet

-0.17

æµħ

-0.15

ekim

-0.15

skyt

-0.14

ÐĴÐ¡

-0.14

 Satisfaction

-0.14

æĿ¡

-0.14

Hut

-0.14

competitive

-0.14

 LoggerFactory

-0.14

POSITIVE LOGITS

loth

0.15

ĥn

0.15

integral

0.15

ÃŃr

0.15

otor

0.15

 beyond

0.15

gren

0.15

 cork

0.14

zen

0.14

Ã±a

0.14

Activations Density 0.220%

instances of overflow or breach leading to physical consequences

No Comments

No Known Activations