INDEX

Explanations

details related to measurement and specifications

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

anzi

-0.16

_prim

-0.15

estate

-0.14

adoo

-0.14

pedo

-0.14

ingly

-0.14

 siguiente

-0.14

Mayor

-0.14

ileo

-0.14

 electrÃ³nico

-0.14

POSITIVE LOGITS

 grandes

0.16

 sentinel

0.16

 volumes

0.15

 exemp

0.15

 memb

0.15

ÃŃveis

0.15

 pesos

0.15

 recipient

0.15

 bloc

0.15

EMPL

0.15

Activations Density 0.013%

details related to measurement and specifications

No Comments

No Known Activations