INDEX

Explanations

terms related to conservatories or conservation 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 @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

agnost

-0.16

cales

-0.15

ãĥ£

-0.15

combe

-0.14

 flag

-0.14

IVATE

-0.13

 Eins

-0.13

sters

-0.13

birds

-0.13

POSITIVE LOGITS

atory

0.28

 Conserv

0.26

ancy

0.23

atoire

0.23

 conserv

0.23

atories

0.21

-cons

0.19

 conservatism

0.19

ational

0.19

ative

0.19

Activations Density 0.004%

terms related to conservatories or conservation in various contexts

No Comments

No Known Activations