INDEX

Explanations

terms related to documents, files, and information exchange in networks

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

termination

-0.15

ufe

-0.15

sequ

-0.14

ulp

-0.14

chu

-0.14

xes

-0.14

Ø§Ø

-0.13

ALLERY

-0.13

dong

-0.13

 hello

-0.13

POSITIVE LOGITS

razier

0.15

AndPassword

0.15

vana

0.14

éĻħ

0.14

èĪ°

0.14

.oracle

0.14

izen

0.13

VR

0.13

à¸Īà¸³

0.13

 management

0.13

Activations Density 0.556%

terms related to documents, files, and information exchange in networks

No Comments

No Known Activations