INDEX

Explanations

references to actions, particularly those associated with performance or activity

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 5-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L5R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.5.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

oons

-0.15

ãģ¡

-0.14

opleft

-0.14

/umd

-0.14

lez

-0.14

ternet

-0.14

Nic

-0.14

vou

-0.13

resident

-0.13

Ð½Ð¸Ñĩ

-0.13

POSITIVE LOGITS

 validationResult

0.17

alsy

0.16

 Fant

0.15

ilyn

0.15

abelle

0.15

kj

0.15

æ½

0.14

 impossible

0.14

.::

0.13

Ð¸Ð½ÑĥÐ²

0.13

Activations Density 0.127%

references to actions, particularly those associated with performance or activity

No Comments

No Known Activations