INDEX

Explanations

phrases and descriptors related to various states of being, particularly in artistic and environmental contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 25-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L25R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.25.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

chten

-0.21

dition

-0.16

á»ĵn

-0.14

åı¦

-0.14

angan

-0.14

detail

-0.14

stants

-0.13

byter

-0.13

ancement

-0.13

agi

-0.13

POSITIVE LOGITS

 complete

0.23

 completely

0.22

 entirely

0.21

 entire

0.21

 entirety

0.20

 Completely

0.20

 Entire

0.19

 partial

0.19

 completo

0.18

 completa

0.18

Activations Density 0.027%

phrases and descriptors related to various states of being, particularly in artistic and environmental contexts

No Comments

No Known Activations