INDEX

Explanations

references to recommendations and comparisons

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 13-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L13R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.13.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

rieg

-0.15

 finally

-0.14

ahl

-0.14

pha

-0.14

insky

-0.14

iri

-0.14

roph

-0.14

iren

-0.14

esc

-0.14

esc

-0.14

POSITIVE LOGITS

 otherwise

0.38

 Otherwise

0.34

Otherwise

0.32

otherwise

0.30

 else

0.29

else

0.26

 OTHERWISE

0.26

 other

0.25

_else

0.25

 jinak

0.25

Activations Density 0.182%

references to recommendations and comparisons

No Comments

No Known Activations