INDEX

Explanations

phrases that indicate references to a company or organization in formal contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

reh

-0.16

 vice

-0.15

arge

-0.15

Off

-0.14

vice

-0.14

izers

-0.14

ency

-0.13

æ¹

-0.13

ream

-0.13

ainers

-0.13

POSITIVE LOGITS

 Pron

0.18

 pron

0.18

 Singular

0.17

 âĨĴ↵↵

0.15

daÅŁ

0.15

-company

0.15

.scalablytyped

0.15

çİĦ

0.14

singular

0.14

olk

0.14

Activations Density 0.008%

phrases that indicate references to a company or organization in formal contexts

No Comments

No Known Activations