INDEX

Explanations

numbers, particularly those representing years or dates

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

kili

-0.16

edir

-0.15

itage

-0.15

omik

-0.15

rosse

-0.15

urette

-0.15

irie

-0.15

ledon

-0.14

à¸¢à¸§

-0.14

htable

-0.14

POSITIVE LOGITS

Ras

0.16

ang

0.16

0.15

ased

0.15

ika

0.15

atum

0.15

uba

0.14

angkan

0.14

ST

0.14

Activations Density 0.011%

numbers, particularly those representing years or dates

No Comments

No Known Activations