INDEX

Explanations

terms related to scientific fields and disciplines

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 20-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L20R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.20.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

ness

-0.17

itude

-0.16

ityEngine

-0.16

Ñķ

-0.16

amento

-0.15

IENT

-0.15

tings

-0.15

sville

-0.15

ability

-0.15

lÄ±k

-0.15

POSITIVE LOGITS

ically

0.18

tainment

0.17

ÚĨÙĩ

0.16

/history

0.16

ALLY

0.16

 department

0.15

Ã«n

0.15

/navigation

0.15

-minded

0.15

/arch

0.15

Activations Density 0.094%

terms related to scientific fields and disciplines

No Comments

No Known Activations