INDEX

Explanations

expressions of emotion and thoughts about 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

Cad

-0.15

igi

-0.15

gio

-0.14

azar

-0.14

apa

-0.14

-0.13

adors

-0.13

ada

-0.13

icity

-0.13

 EntityState

-0.13

POSITIVE LOGITS

Ã¼zel

0.15

 ÙħØ²

0.14

dsl

0.14

arl

0.13

alue

0.13

mÃ®

0.13

èĥİ

0.13

OOT

0.13

Ã¼ssen

0.13

meld

0.13

Activations Density 0.310%

expressions of emotion and thoughts about personal experiences

No Comments

No Known Activations