INDEX

Explanations

references to specific titles or characters in media

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

ÐµÑĢÐ¾

-0.15

istrovstvÃŃ

-0.15

 outlier

-0.14

.Sdk

-0.14

--)

-0.14

à¸°

-0.14

zion

-0.13

èĩ£

-0.13

Scores

-0.13

.configure

-0.13

POSITIVE LOGITS

emean

0.17

CS

0.16

ÑĥÐ¼Ð°

0.15

 Tommy

0.14

Con

0.14

Cal

0.14

 Kostenlose

0.13

ouse

0.13

umu

0.13

Activations Density 0.170%

references to specific titles or characters in media

No Comments

No Known Activations