INDEX

Explanations

questions directed at the reader about expertise and personal experiences

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

ulan

-0.16

__("

-0.14

ekt

-0.14

\Migration

-0.14

hood

-0.14

 Rubin

-0.14

 Guidance

-0.14

.prototype

-0.13

 adher

-0.13

wed

-0.13

POSITIVE LOGITS

 çŁ

0.16

ayne

0.14

ogue

0.14

dou

0.14

idental

0.14

 Duffy

0.14

.CurrentRow

0.14

lobals

0.14

 your

0.14

 Suff

0.13

Activations Density 0.151%

questions directed at the reader about expertise and personal experiences

No Comments

No Known Activations