INDEX

Explanations

phrases that emphasize assistance or offers of help to the reader or listener

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

èĥİ

-0.16

Sez

-0.14

Î³ÎºÎ±

-0.14

£¼

-0.14

 Surround

-0.14

ãģĤãģĴ

-0.14

(åľŁ

-0.14

ingen

-0.14

issance

-0.14

geois

-0.14

POSITIVE LOGITS

obus

0.17

}elseif

0.17

/us

0.17

 transition

0.15

-transition

0.14

lon

0.14

transition

0.14

yan

0.14

.jp

0.14

SearchParams

0.14

Activations Density 0.069%

phrases that emphasize assistance or offers of help to the reader or listener

No Comments

No Known Activations