INDEX

Explanations

terms related to "wide" formations or positions in sports

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

idan

-0.15

ÙħØ²

-0.15

ieri

-0.15

 Schwarz

-0.14

amax

-0.14

ocale

-0.14

 dair

-0.14

tro

-0.14

spb

-0.14

oproject

-0.13

POSITIVE LOGITS

outs

0.28

out

0.24

 receiver

0.23

 receivers

0.21

oute

0.19

-outs

0.18

 Receiver

0.17

_receiver

0.17

 Route

0.17

ourt

0.17

Activations Density 0.004%

terms related to "wide" formations or positions in sports

No Comments

No Known Activations