INDEX

Explanations

elements related to memorable experiences or lasting impressions in a community context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

__

-0.15

 rounded

-0.14

 round

-0.14

 saint

-0.14

 completes

-0.14

inity

-0.14

 dialect

-0.14

 complet

-0.13

mys

-0.13

otto

-0.13

POSITIVE LOGITS

 memory

0.25

memory

0.24

/memory

0.22

-memory

0.22

 MEMORY

0.22

 Memory

0.21

.memory

0.21

MEMORY

0.21

 memories

0.21

lod

0.21

Activations Density 0.118%

elements related to memorable experiences or lasting impressions in a community context

No Comments

No Known Activations