INDEX

Explanations

mentions of technology and product features

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

esModule

-0.17

hlas

-0.15

deaux

-0.15

oli

-0.14

èĥİ

-0.14

abad

-0.14

/pi

-0.14

omorphic

-0.14

jiang

-0.14

ÙİØ¯

-0.14

POSITIVE LOGITS

sur

0.23

 Rolling

0.22

Sur

0.21

Roll

0.21

Sur

0.20

roll

0.20

 Roll

0.20

SUR

0.20

 roll

0.19

ROLL

0.19

Activations Density 0.029%

mentions of technology and product features

No Comments

No Known Activations

mentions of technology and product features

No Comments

No Known Activations