INDEX

Explanations

numerical information or dates in a text

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

orum

-0.16

ohn

-0.14

 Campo

-0.14

ega

-0.14

bing

-0.14

ules

-0.14

otta

-0.13

loor

-0.13

jan

-0.13

iren

-0.13

POSITIVE LOGITS

(iOS

0.17

mbH

0.16

edback

0.14

anÃ©

0.14

bose

0.14

rita

0.14

 licensors

0.14

ride

0.13

OfDay

0.13

Activations Density 0.130%

numerical information or dates in a text

No Comments

No Known Activations