INDEX

Explanations

occurrences of the word "Si" 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 @ 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

leo

-0.17

chein

-0.16

bih

-0.16

ÑĨÐµÑĢ

-0.16

 Frog

-0.16

pez

-0.15

 frog

-0.15

 appare

-0.15

orio

-0.15

iye

-0.14

POSITIVE LOGITS

ames

0.28

enna

0.26

emens

0.25

erra

0.23

oux

0.23

empre

0.23

esta

0.20

erras

0.19

ERR

0.18

houette

0.18

Activations Density 0.009%

occurrences of the word "Si" in various contexts

No Comments

No Known Activations