INDEX

Explanations

possessive forms of names or titles, indicating ownership or association

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

æ¢

-0.15

icens

-0.15

prd

-0.15

tant

-0.14

rant

-0.14

ightly

-0.14

Äĩ

-0.14

fad

-0.14

æĥħ

-0.14

 Neville

-0.13

POSITIVE LOGITS

raison

0.15

ATEG

0.15

 kami

0.14

klÃ¤

0.14

keley

0.14

.GroupLayout

0.14

elik

0.14

ÐµÑĢÐ°ÑĤÑĥÑĢÐ°

0.14

ksi

0.13

ehler

0.13

Activations Density 0.083%

possessive forms of names or titles, indicating ownership or association

No Comments

No Known Activations

possessive forms of names or titles, indicating ownership or association

No Comments

No Known Activations