INDEX

Explanations

possessive forms related to people's attributes or belongings

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 2-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L2R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.2.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

ynos

-0.19

Pink

-0.15

ney

-0.14

 gang

-0.14

 Pink

-0.14

mf

-0.14

ÑĢÐ¸Ð·

-0.14

 Watts

-0.14

esin

-0.13

ising

-0.13

POSITIVE LOGITS

/docs

0.15

rale

0.14

lest

0.14

 rele

0.14

elho

0.14

.Scheme

0.14

ableView

0.14

ÏģÏīÎ½

0.14

 Shuffle

0.13

 komp

0.13

Activations Density 0.097%

possessive forms related to people's attributes or belongings

No Comments

No Known Activations