INDEX

Explanations

terms related to software components and their interactions within a system

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

eka

-0.19

sWith

-0.18

eur

-0.18

ningen

-0.18

ek

-0.17

ery

-0.17

spacer

-0.17

ti

-0.17

nid

-0.17

eki

-0.17

POSITIVE LOGITS

hip

0.33

-upper

0.26

ship

0.26

/editor

0.23

SHIP

0.22

hips

0.22

/compiler

0.22

ial

0.21

/loader

0.21

 beware

0.21

Activations Density 0.534%

terms related to software components and their interactions within a system

No Comments

No Known Activations