INDEX

Explanations

phrases and actions related to the purpose or intended use of items or services

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 23-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L23R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.23.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

 Ð¿ÑĢÐ¸Ð¿

-0.15

à¤¾à¤Ł

-0.14

eam

-0.14

 GOODMAN

-0.14

oj

-0.14

acci

-0.14

æĦı

-0.14

ALES

-0.13

 Buying

-0.13

POSITIVE LOGITS

use

0.57

use

0.41

Use

0.37

Use

0.37

 inclusion

0.36

.use

0.36

_use

0.35

uso

0.34

 consumption

0.34

-use

0.31

Activations Density 0.157%

phrases and actions related to the purpose or intended use of items or services

No Comments

No Known Activations