INDEX

Explanations

terms related to challenges and feasibility in various contexts

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

ãģĵãĤĵãģ«ãģ¡ãģ¯

-0.15

ongan

-0.15

ONGL

-0.14

Because

-0.14

Descending

-0.14

 ofType

-0.14

mrt

-0.14

 porque

-0.14

 voks

-0.14

POSITIVE LOGITS

du

0.29

-d

0.25

duo

0.24

do

0.23

du

0.23

ue

0.23

 owning

0.21

dov

0.20

 clue

0.19

Activations Density 0.249%

terms related to challenges and feasibility in various contexts

No Comments

No Known Activations