INDEX

Explanations

instances of the word "select" and its variations in the context of choosing or identifying items or options

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

lox

-0.18

zon

-0.17

arguments

-0.16

olf

-0.15

oret

-0.15

oca

-0.14

 Arguments

-0.14

ayd

-0.14

.shop

-0.14

Ã¼yÃ¼k

-0.14

POSITIVE LOGITS

çŃĴ

0.16

.preview

0.15

/select

0.14

fi

0.14

 Hodg

0.14

 Voll

0.14

éĹ

0.14

errupted

0.14

ylko

0.14

corner

0.14

Activations Density 0.019%

instances of the word "select" and its variations in the context of choosing or identifying items or options

No Comments

No Known Activations