INDEX

Explanations

punctuation marks and their usage in written text

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

inkle

-0.15

Forge

-0.15

vit

-0.14

bern

-0.14

Seeder

-0.14

lyph

-0.14

ÑĪÐµ

-0.14

ForObject

-0.14

Î¿ÏįÏĤ

-0.14

ota

-0.13

POSITIVE LOGITS

 Fund

0.15

iola

0.15

0.14

iya

0.14

phan

0.14

BarButton

0.13

 Carson

0.13

.train

0.13

 Helm

0.13

anship

0.13

Activations Density 0.048%

punctuation marks and their usage in written text

No Comments

No Known Activations