INDEX

Explanations

references to publications or sources, particularly historical and statistical content

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L8R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.8.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

.lesson

-0.16

vore

-0.15

ymi

-0.14

Äģn

-0.14

ä¿

-0.14

ãģıãĤī

-0.14

inn

-0.13

etails

-0.13

.scalablytyped

-0.13

Î±Î¹

-0.13

POSITIVE LOGITS

uber

0.17

 sensitive

0.15

ebek

0.14

sur

0.14

DRV

0.14

AREST

0.14

-sur

0.13

fec

0.13

 ends

0.13

estar

0.13

Activations Density 0.163%

references to publications or sources, particularly historical and statistical content

No Comments

No Known Activations