INDEX

Explanations

references to people or groups of people and their activities or information about them

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ÑħÐ¾Ð´Ð¸Ð¼

-0.16

layÄ±n

-0.15

ifr

-0.14

quir

-0.14

.vertx

-0.14

âĺħâĺħ

-0.14

utex

-0.14

.crt

-0.13

itized

-0.13

anth

-0.13

POSITIVE LOGITS

å·¨

0.17

zik

0.16

meg

0.14

 ones

0.14

 excer

0.14

_SCL

0.14

pest

0.13

 Mint

0.13

onta

0.13

Activations Density 0.442%

references to people or groups of people and their activities or information about them

No Comments

No Known Activations