INDEX

Explanations

discussions about influential individuals and their performances or roles

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

ubits

-0.16

abet

-0.15

rame

-0.15

.prod

-0.14

cig

-0.14

oid

-0.14

 Rock

-0.14

 Foster

-0.14

 rocker

-0.14

POSITIVE LOGITS

 continue

0.21

 ê³ĦìĨį

0.20

continue

0.20

 increasingly

0.19

 consistently

0.19

 continues

0.19

 continued

0.19

 continuing

0.17

continued

0.17

 lately

0.16

Activations Density 0.261%

discussions about influential individuals and their performances or roles

No Comments

No Known Activations