INDEX

Explanations

references to formatting issues or technical problems in document editing

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

 redirects

-0.14

 redirect

-0.14

lfw

-0.14

å¹³åı°

-0.14

_widget

-0.14

tar

-0.14

iri

-0.14

å¯«

-0.14

owl

-0.13

åį

-0.13

POSITIVE LOGITS

 commands

0.22

 Format

0.20

 Styles

0.19

 Properties

0.19

 Object

0.18

 backstage

0.18

 Inspector

0.18

_Format

0.18

 Ribbon

0.18

 command

0.17

Activations Density 0.028%

references to formatting issues or technical problems in document editing

No Comments

No Known Activations