INDEX

Explanations

dialogue prompts and question formats within conversations

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

ereum

-0.18

ott

-0.17

edin

-0.15

isu

-0.14

Å¡k

-0.14

throp

-0.14

incare

-0.14

oure

-0.13

ques

-0.13

lett

-0.13

POSITIVE LOGITS

eza

0.13

Latch

0.13

ocht

0.13

Rit

0.13

ÎŃÎ»

0.13

_undo

0.13

nod

0.13

chie

0.13

Ð½Ð¾Ð´

0.13

 fila

0.12

Activations Density 0.028%

dialogue prompts and question formats within conversations

No Comments

No Known Activations