INDEX

Explanations

proper nouns

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

rowse

-0.16

.appspot

-0.16

ideo

-0.15

RenderingContext

-0.15

ippets

-0.15

activex

-0.15

 Ä°li

-0.15

SON

-0.14

isphere

-0.14

otope

-0.14

POSITIVE LOGITS

 Boys

0.16

con

0.15

 boys

0.15

 Draco

0.15

ç®±

0.15

/B

0.14

 Persistent

0.14

\Context

0.14

 overnight

0.14

 Babies

0.14

Activations Density 0.105%

proper nouns

No Comments

No Known Activations