INDEX

Explanations

phrases related to quality and superiority, often indicating the best options available

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 17-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L17R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.17.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

cern

-0.14

Ã¼Ã§Ã¼k

-0.14

Dut

-0.14

elt

-0.14

-0.13

Ð¾ÑĢÐ¾

-0.13

antu

-0.13

 favorites

-0.13

 hero

-0.13

ÐµÑĢÐ¾

-0.13

POSITIVE LOGITS

-selling

0.19

-known

0.19

seller

0.17

/fast

0.17

lest

0.16

ever

0.16

-case

0.16

owing

0.15

Ø§Ø¨Ø±

0.15

-looking

0.15

Activations Density 0.048%

phrases related to quality and superiority, often indicating the best options available

No Comments

No Known Activations