INDEX

Explanations

phrases indicating emotional states or actions related to support and care

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

urre

-0.15

Siz

-0.15

_pres

-0.15

éº»

-0.14

nnen

-0.14

ifo

-0.14

pray

-0.13

igious

-0.13

POSITIVE LOGITS

 next

0.22

 Next

0.21

 circ

0.21

 Falls

0.20

next

0.19

 Fell

0.18

NEXT

0.18

 fall

0.18

èĲ½

0.17

 fell

0.17

Activations Density 0.044%

phrases indicating emotional states or actions related to support and care

No Comments

No Known Activations