INDEX

Explanations

various non-zero numerical values and unique identifiers in a technical context

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

Ð´Ð°Ð¿

-0.16

 Swinger

-0.16

rio

-0.14

ÑĨÐ¸ÑĤ

-0.14

ubit

-0.14

asley

-0.14

 McInt

-0.14

uka

-0.14

rani

-0.14

metro

-0.14

POSITIVE LOGITS

agement

0.15

cube

0.14

amble

0.14

aign

0.14

routeParams

0.13

 Salv

0.13

atest

0.13

ë¹Ħ

0.13

iped

0.13

GNU

0.13

Activations Density 0.034%

various non-zero numerical values and unique identifiers in a technical context

No Comments

No Known Activations