INDEX

Explanations

phrases that quantify or refer to the concept of "many."

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

 most

-0.17

æľĢ

-0.15

ambda

-0.14

_iff

-0.14

ãĤĪãģı

-0.14

opic

-0.13

losures

-0.13

 eines

-0.13

iya

-0.13

ibe

-0.13

POSITIVE LOGITS

 ways

0.34

 reasons

0.30

 reason

0.27

reason

0.24

 Ways

0.24

 places

0.24

 Reasons

0.21

Reason

0.20

ways

0.20

 occasions

0.20

Activations Density 0.136%

phrases that quantify or refer to the concept of "many."

No Comments

No Known Activations