INDEX

Explanations

numerical values and sizes in code or technical documentation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 7-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L7R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.7.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

/Library

-0.15

auf

-0.15

vements

-0.14

½Ķ

-0.14

*>*

-0.14

cko

-0.14

-0.13

ban

-0.13

 Freed

-0.13

.bam

-0.13

POSITIVE LOGITS

AuthProvider

0.16

ABCDEFGHIJKLMNOP

0.16

aeda

0.16

quist

0.16

iliz

0.14

Ã©

0.14

quan

0.14

ServiceProvider

0.14

hei

0.14

ritch

0.14

Activations Density 0.111%

numerical values and sizes in code or technical documentation

No Comments

No Known Activations