INDEX

Explanations

references to specific names or entities, likely within a context of sports teams or competitions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

onda

-0.17

rov

-0.16

ays

-0.15

 kims

-0.14

aÃ±

-0.14

ahl

-0.14

.ResumeLayout

-0.14

cke

-0.14

ECH

-0.13

@update

-0.13

POSITIVE LOGITS

 captain

0.22

 yellow

0.21

 capt

0.20

 Yellow

0.20

Yellow

0.19

 captains

0.19

_yellow

0.19

(c

0.18

undle

0.18

capt

0.17

Activations Density 0.033%

references to specific names or entities, likely within a context of sports teams or competitions

No Comments

No Known Activations