INDEX

Explanations

phrases that initiate conditional statements or exclusions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 5-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L5R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.5.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

usk

-0.16

jee

-0.15

egan

-0.14

kan

-0.14

aight

-0.14

dech

-0.14

edb

-0.14

STYPE

-0.13

ffer

-0.13

POSITIVE LOGITS

fed

0.16

æĹ

0.15

.Compiler

0.15

æĥ

0.14

ocha

0.14

Î»Î¿Ïħ

0.14

ucu

0.13

 necessary

0.13

recht

0.13

 toll

0.13

Activations Density 0.011%

phrases that initiate conditional statements or exclusions

No Comments

No Known Activations