INDEX

Explanations

references to specific years or timeframes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

 earlier

-0.18

 recently

-0.17

 lately

-0.17

 later

-0.16

ayer

-0.16

 recent

-0.15

 currently

-0.15

specifier

-0.15

recent

-0.15

lest

-0.15

POSITIVE LOGITS

 ÙħÛĮÙĦØ§Ø¯ÛĮ

0.17

elyn

0.17

strup

0.17

ByExample

0.16

ovo

0.15

ingly

0.15

eon

0.14

irts

0.14

etchup

0.14

/MPL

0.14

Activations Density 0.025%

references to specific years or timeframes

No Comments

No Known Activations