INDEX

Explanations

numerical or tabular data representations

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

éİ®

-0.14

 openFileDialog

-0.14

awan

-0.14

lobal

-0.14

ebo

-0.13

 Aware

-0.13

Peb

-0.13

 Consort

-0.13

lein

-0.13

pec

-0.13

POSITIVE LOGITS

arro

0.17

idence

0.14

Lau

0.14

unning

0.14

à¤¼à¤°

0.13

 coff

0.13

Law

0.13

quo

0.13

airie

0.12

ÙĨÚ¯

0.12

Activations Density 0.006%

numerical or tabular data representations

No Comments

No Known Activations