INDEX

Explanations

key terms and phrases that indicate significant actions, quantities, and relationships in the context of analysis or research

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 13-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L13R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.13.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

awks

-0.16

ãĤ¯ãĥª

-0.15

 hlav

-0.15

xies

-0.14

á»ı

-0.14

reesome

-0.14

BITS

-0.14

ãĤ¢ãĥ³

-0.14

erne

-0.14

VERS

-0.13

POSITIVE LOGITS

 PROVIDED

0.18

 Provided

0.16

Ð¾ÐºÐ¾Ð½

0.16

atatype

0.15

ÐĿÐ¯

0.15

 provided

0.15

 touch

0.15

urons

0.15

Touch

0.14

cul

0.14

Activations Density 0.013%

key terms and phrases that indicate significant actions, quantities, and relationships in the context of analysis or research

No Comments

No Known Activations