INDEX

Explanations

technical instructions related to error messages and user actions

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

modulo

-0.16

 nuest

-0.16

uye

-0.16

 ragaz

-0.15

 siguientes

-0.15

Venta

-0.15

 primero

-0.15

 synt

-0.14

 coll

-0.14

POSITIVE LOGITS

 mais

0.20

Ã´n

0.19

inha

0.18

Ãªm

0.17

lou

0.16

inho

0.16

ente

0.15

weis

0.15

æµª

0.15

ou

0.15

Activations Density 0.125%

technical instructions related to error messages and user actions

No Comments

No Known Activations