INDEX

Explanations

references to educational and professional advancement

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

ãĥ«ãĤ¯

-0.15

uÄį

-0.15

viar

-0.14

ugi

-0.14

lj

-0.14

uje

-0.14

okers

-0.14

à¥Įà¤²

-0.14

uria

-0.14

dogs

-0.14

POSITIVE LOGITS

iban

0.19

iper

0.15

ENTA

0.14

emony

0.14

borg

0.14

Truthy

0.14

agi

0.14

mitter

0.14

annon

0.13

ips

0.13

Activations Density 0.377%

references to educational and professional advancement

No Comments

No Known Activations