INDEX

Explanations

elements related to visual and contextual descriptions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 23-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L23R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.23.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

ulet

-0.15

ouble

-0.14

æĶ¾åľ¨

-0.14

éº

-0.14

 tucked

-0.13

 stretched

-0.13

arest

-0.13

_extended

-0.13

è¾°

-0.13

apter

-0.13

POSITIVE LOGITS

 filled

0.50

filled

0.38

 full

0.38

Filled

0.32

 pepper

0.32

 litter

0.31

 dominated

0.31

 llen

0.29

 fill

0.29

 FULL

0.28

Activations Density 0.301%

elements related to visual and contextual descriptions

No Comments

No Known Activations