INDEX

Explanations

various forms of content and actions, particularly related to configuration and documentation

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

LATED

-0.14

à¥ģà¤¨

-0.14

Ī

-0.13

 foot

-0.13

 fortress

-0.13

SetValue

-0.13

->___

-0.13

 Footer

-0.13

izens

-0.12

 recession

-0.12

POSITIVE LOGITS

umo

0.19

irth

0.15

ilo

0.15

0.14

yr

0.14

cel

0.14

section

0.14

neh

0.13

ule

0.13

ky

0.13

Activations Density 0.030%

various forms of content and actions, particularly related to configuration and documentation

No Comments

No Known Activations