INDEX

Explanations

occurrences of the term "push" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

ignon

-0.17

agle

-0.17

utsch

-0.16

adle

-0.16

sein

-0.16

Ð¸ÑĩÐ½Ð¾

-0.15

earch

-0.15

Ð°ÑĢÑĩ

-0.15

igor

-0.15

Ø³Ø¯

-0.14

POSITIVE LOGITS

(push

0.27

.Push

0.24

 push

0.24

rod

0.23

-push

0.23

pull

0.22

cart

0.21

pin

0.21

 Push

0.21

 pushing

0.21

Activations Density 0.023%

occurrences of the term "push" in various contexts

No Comments

No Known Activations