INDEX

Explanations

terms related to sports outcomes, scores, and player performances

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

uard

-0.19

ijo

-0.15

 ******************************************************************************↵

-0.14

ãĥĬãĥ¼

-0.14

guard

-0.14

bris

-0.14

odem

-0.13

ilim

-0.13

toPromise

-0.13

Ī

-0.13

POSITIVE LOGITS

/-

0.21

 final

0.19

âĢ³

0.19

 decision

0.18

 lead

0.18

Â½

0.17

 margin

0.17

win

0.16

in

0.15

 after

0.15

Activations Density 0.030%

terms related to sports outcomes, scores, and player performances

No Comments

No Known Activations