INDEX

Explanations

references to individuals named Andrew

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 15-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L15R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.15.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

ernen

-0.18

ucch

-0.17

erap

-0.17

ÌĨ

-0.16

eru

-0.15

erah

-0.15

 Needle

-0.15

yun

-0.15

omi

-0.15

chartInstance

-0.14

POSITIVE LOGITS

ries

0.17

ium

0.17

kes

0.16

edor

0.16

å®¶çļĦ

0.16

lings

0.16

verse

0.15

orld

0.15

ery

0.15

Activations Density 0.013%

references to individuals named Andrew

No Comments

No Known Activations