INDEX

Explanations

terms related to racing and competition

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

çĿ£

-0.18

-0.17

lie

-0.16

sson

-0.15

 kter

-0.15

ths

-0.15

estruct

-0.15

shire

-0.15

ness

-0.14

POSITIVE LOGITS

horse

0.19

erb

0.17

TokenType

0.17

erd

0.16

erp

0.15

.debian

0.15

LAN

0.14

erin

0.14

ableObject

0.14

MON

0.14

Activations Density 0.027%

terms related to racing and competition

No Comments

No Known Activations