INDEX

Explanations

phrases that express relationships or connections between concepts

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

-0.15

rok

-0.15

mente

-0.14

anko

-0.14

TRS

-0.14

astle

-0.14

hood

-0.14

dog

-0.14

 deadlock

-0.14

iam

-0.14

POSITIVE LOGITS

irect

0.16

adoo

0.16

 Clothes

0.16

 diret

0.15

 DIRECT

0.15

edor

0.15

 ayar

0.14

chner

0.14

estro

0.14

DIRECT

0.14

Activations Density 0.016%

phrases that express relationships or connections between concepts

No Comments

No Known Activations