INDEX

Explanations

actions or commands related to interacting with a device or machine

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 7-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L7R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.7.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

ourcem

-0.17

Orm

-0.16

Injector

-0.15

morgan

-0.14

anche

-0.14

reu

-0.14

bette

-0.14

slashes

-0.14

enco

-0.14

Ã©lÃ©

-0.14

POSITIVE LOGITS

hold

0.14

lickr

0.14

hom

0.14

 licensors

0.14

 Armour

0.14

 Barcl

0.14

ungle

0.13

ibly

0.13

hole

0.13

holding

0.13

Activations Density 0.026%

actions or commands related to interacting with a device or machine

No Comments

No Known Activations