INDEX

Explanations

terms related to collaboration and community roles in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

utos

-0.15

 Ð´Ð¾ÑĢÐ¾Ð¶

-0.14

assis

-0.14

ortal

-0.14

leness

-0.14

ÑĨÐµÐ¿

-0.14

NOWLED

-0.14

 Ð¾ÑģÐ¾Ð±Ð»Ð¸Ð²Ð¾

-0.14

 Ð·Ð½Ð°ÑĩÐ¸ÑĤ

-0.13

enburg

-0.13

POSITIVE LOGITS

 happens

0.24

 happen

0.22

 belonged

0.21

0.19

 happened

0.19

 besides

0.19

 include

0.18

 happening

0.18

none

0.18

 belong

0.18

Activations Density 0.085%

terms related to collaboration and community roles in various contexts

No Comments

No Known Activations