INDEX

Explanations

references to durations or counts in a sequence, especially concerning terms or events

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

æİĮ

-0.15

ucker

-0.14

 jadx

-0.14

åı¥

-0.14

occo

-0.14

 toll

-0.14

erras

-0.14

achts

-0.14

alg

-0.13

ÙĪÙĨÙĬ

-0.13

POSITIVE LOGITS

affer

0.16

PCA

0.16

ingham

0.15

ardy

0.15

erten

0.15

polator

0.15

å¿Ĺ

0.15

arian

0.14

ukan

0.14

unte

0.14

Activations Density 0.161%

references to durations or counts in a sequence, especially concerning terms or events

No Comments

No Known Activations

references to durations or counts in a sequence, especially concerning terms or events

No Comments

No Known Activations