INDEX

Explanations

conjunctions and descriptors related to various facets of personal and professional identities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 20-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L20R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.20.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

aux

-0.17

ieu

-0.15

ooter

-0.14

$?

-0.14

lee

-0.14

 rost

-0.14

tery

-0.14

edish

-0.13

argent

-0.13

 Legendary

-0.13

POSITIVE LOGITS

 member

0.22

 former

0.22

 erst

0.21

 author

0.17

 part

0.17

 frequent

0.16

member

0.15

requent

0.15

 overall

0.15

 friend

0.15

Activations Density 0.067%

conjunctions and descriptors related to various facets of personal and professional identities

No Comments

No Known Activations