INDEX

Explanations

numerical data, particularly related to measurements or quantities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

illis

-0.18

iliz

-0.17

gregator

-0.15

entiful

-0.15

igram

-0.14

imiters

-0.14

cly

-0.14

glomer

-0.14

ofday

-0.13

POSITIVE LOGITS

alcon

0.17

uet

0.16

ÑģÑĤÐ°Ð½

0.15

VEN

0.15

Eig

0.15

.heroku

0.15

 Ferm

0.14

CLK

0.13

 Moff

0.13

StringBuilder

0.13

Activations Density 0.004%

numerical data, particularly related to measurements or quantities

No Comments

No Known Activations