INDEX

Explanations

references to different types of space and spatial concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

/respond

-0.17

spot

-0.15

ç´ł

-0.15

abyrin

-0.15

aceae

-0.15

_species

-0.14

ãĥĲãĥ¼

-0.14

rex

-0.14

 ÑģÐºÐ»Ð°Ð´Ñĥ

-0.14

ne

-0.14

POSITIVE LOGITS

-temp

0.20

/time

0.20

yonel

0.19

uits

0.18

walk

0.18

-time

0.18

walking

0.18

-Time

0.17

 heater

0.17

hips

0.17

Activations Density 0.051%

references to different types of space and spatial concepts

No Comments

No Known Activations