INDEX

Explanations

references to vehicle specifications and racing history

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

ornings

-0.17

 Mech

-0.15

aki

-0.15

lobal

-0.15

CORE

-0.14

ucz

-0.14

_rsp

-0.14

statt

-0.14

 Airbus

-0.14

 vess

-0.14

POSITIVE LOGITS

 Sport

0.17

 production

0.16

 sport

0.16

perimental

0.16

 sporting

0.16

 body

0.15

 chassis

0.15

aight

0.15

 Sports

0.15

 Production

0.15

Activations Density 0.038%

references to vehicle specifications and racing history

No Comments

No Known Activations