INDEX

Explanations

statistical data and percentage representations

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

-placement

-0.16

 elsewhere

-0.16

olog

-0.15

aled

-0.15

.scalablytyped

-0.15

 Placement

-0.15

 serialVersionUID

-0.14

Ãłn

-0.14

irts

-0.14

POSITIVE LOGITS

ingo

0.16

alter

0.15

ENAME

0.15

inth

0.15

rens

0.14

 novelty

0.14

åĿĬ

0.14

ifle

0.14

æŁľ

0.14

locker

0.14

Activations Density 0.058%

statistical data and percentage representations

No Comments

No Known Activations