INDEX

Explanations

time indicators and schedules

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

dsl

-0.15

sembly

-0.14

 ÐºÐ¾Ð²

-0.14

cko

-0.14

ARRIER

-0.14

imizer

-0.13

adows

-0.13

ContentAlignment

-0.13

xon

-0.13

engo

-0.13

POSITIVE LOGITS

Tester

0.14

 Garten

0.14

isle

0.13

ucci

0.13

Decorator

0.13

iag

0.13

erville

0.13

 presentViewController

0.13

 nighttime

0.13

ÏģÎŃ

0.13

Activations Density 0.044%

time indicators and schedules

No Comments

No Known Activations