INDEX

Explanations

specific identifiers or labels related to scientific publications or references

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

yonel

-0.16

chwitz

-0.15

kenin

-0.15

 Ekon

-0.14

ercial

-0.14

thora

-0.14

PCS

-0.14

ispens

-0.14

redo

-0.14

 Pharmac

-0.13

POSITIVE LOGITS

 personal

0.17

æİĴ

0.16

.cd

0.15

 plain

0.15

Dod

0.15

Jet

0.15

éĮ²

0.15

 pers

0.15

ÑģÑĤÐ¸

0.15

bane

0.14

Activations Density 0.129%

specific identifiers or labels related to scientific publications or references

No Comments

No Known Activations