INDEX

Explanations

symbols and formats commonly associated with contact information or numerical data

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

áº¯m

-0.15

inae

-0.15

ildo

-0.15

.AddListener

-0.15

ulet

-0.15

ecs

-0.14

Î¼ÎŃÏĤ

-0.14

nul

-0.14

èªĮ

-0.14

isson

-0.14

POSITIVE LOGITS

_macros

0.17

zahl

0.15

 Imag

0.15

 MÃ£

0.14

Cop

0.14

 Gregg

0.14

svn

0.14

 Ø³Ùģ

0.13

çłģ

0.13

Mak

0.13

Activations Density 0.013%

symbols and formats commonly associated with contact information or numerical data

No Comments

No Known Activations