INDEX

Explanations

the presence of specific articles, pronouns, and referential terms that indicate the subject, object, or action in a narrative context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 20-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L20R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.20.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

-0.15

 either

-0.14

 inplace

-0.14

 daily

-0.14

spy

-0.14

assin

-0.13

QN

-0.13

sten

-0.13

 intact

-0.13

.vn

-0.13

POSITIVE LOGITS

 actions

0.20

 Ð´ÐµÐ¹ÑģÑĤÐ²Ð¸Ñı

0.19

 feat

0.19

è¡Įåĭķ

0.18

actions

0.17

 effort

0.17

 viá»ĩc

0.17

 action

0.17

ialog

0.16

 activity

0.16

Activations Density 0.006%

the presence of specific articles, pronouns, and referential terms that indicate the subject, object, or action in a narrative context

No Comments

No Known Activations