INDEX

Explanations

quantifiable metrics and relationships within a network or group

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

ingles

-0.18

ÅĻet

-0.17

ichen

-0.15

quine

-0.14

ÙĤÙĩ

-0.14

Ð¾Ð²Ð¸

-0.14

 doma

-0.14

gesi

-0.14

eks

-0.14

esco

-0.14

POSITIVE LOGITS

two

0.26

 three

0.24

 across

0.24

 several

0.23

 multiple

0.23

 between

0.21

 various

0.20

two

0.20

all

0.18

Across

0.18

Activations Density 0.159%

quantifiable metrics and relationships within a network or group

No Comments

No Known Activations