INDEX

Explanations

references to business entities or organizations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

actionDate

-0.13

FIG

-0.13

iability

-0.13

enge

-0.12

Bas

-0.12

/dashboard

-0.12

èĬĤ

-0.12

/embed

-0.12

blem

-0.12

okit

-0.12

POSITIVE LOGITS

LOPT

0.14

0.13

ÑĢÑĸÐ·

0.13

Ãłn

0.13

 olursa

0.13

=int

0.13

Activations Density 0.292%

references to business entities or organizations

No Comments

No Known Activations