INDEX

Explanations

various types of brackets and braces, indicating structured elements such as code or lists

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 2-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L2R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.2.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

 NavParams

-0.15

å¾Ģ

-0.15

ÑıÑīÐ¸Ñħ

-0.14

elow

-0.14

zin

-0.14

zon

-0.14

 Mell

-0.13

elman

-0.13

 Stranger

-0.13

.argsort

-0.13

POSITIVE LOGITS

UNS

0.17

starter

0.15

.ids

0.15

dam

0.14

-host

0.14

 Propel

0.14

 acad

0.14

 host

0.14

-ac

0.14

 Academic

0.14

Activations Density 0.010%

various types of brackets and braces, indicating structured elements such as code or lists

No Comments

No Known Activations